Fundamentals of the Gut for Capsule Engineers

This is a repository copy of Fundamentals of the gut for capsule engineers.

White Rose Research Online URL for this paper: https://eprints.whiterose.ac.uk/164293/

Version: Accepted Version

Article: Barducci, L, Norton, J, Sarker, S et al. (4 more authors) (2020) Fundamentals of the gut for capsule engineers. Progress in Biomedical Engineering. ISSN 2516-1091 https://doi.org/10.1088/2516-1091/abab4c

Reuse This article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can’t change the article in any way or use it commercially. More information and the full terms of the licence here: https://creativecommons.org/licenses/

Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [email protected] including the URL of the record and the reason for the withdrawal request.

[email protected] https://eprints.whiterose.ac.uk/

1 Fundamentals of the Gut for Capsule Engineers

2 Lavinia Barducci*, Joseph C. Norton, Sunandita Sarker, Sayeed Mohammed, Ryan Jones, Pietro

3 Valdastri, Benjamin S. Terry

5 Lavinia Barducci*, M.Sc in Robotics and Automation Engineering, STORM Lab UK, Institute

6 of Robotics, School of Electronic and Electrical Engineering, University of Leeds, Woodhouse

7 Lane, Leeds, UK. LS2 9JT. [email protected]

8 Joseph C. Norton, PhD Mechanical Engineering, STORM Lab UK, Institute of Robotics, School

9 of Electronic and Electrical Engineering, University of Leeds, Woodhouse Lane, Leeds, UK. LS2

10 9JT. [email protected]

11 Sunandita Sarker, PhD candidate, Mechanical engineering and applied mechanics, Terry

12 Research Lab, Department of Mechanical and Materials Engineering, University of Nebraska

13 Lincoln, [email protected], Postal address: 360 Walter Scott Engineering Center,

14 University of Nebraska, Lincoln, Nebraska, 68588, USA

15 Sayeed Mohammed, B.Sc in Mechanical Engineering, Mechanical Engineering, University of

16 Nebraska Lincoln, [email protected], Grand Valley State University, 49504

17 Ryan Jones, BEng, VP, Corporate Development, Progenity Inc., 5230 S. State Road, Ann Arbor,

18 MI 48108, [email protected].

19 Pietro Valdastri, PhD Electronic Engineering, STORM Lab UK, Institute of Robotics, School of

20 Electronic and Electrical Engineering, University of Leeds, Woodhouse Lane, Leeds, UK. LS2

21 9JT, [email protected]

22 Benjamin S. Terry, Ph.D., Department of Mechanical and Materials Engineering, University of

23 Nebraska Lincoln, [email protected], Postal address: UNL, 317.4B Nebraska Hall, Lincoln,

24 Nebraska, 68588, USA, No financial arrangements or conflict of interest.

26 *Corresponding author

28 Figure 1: Anatomy of the digestive system, showing all organs...... 13 29 Figure 2: The functional regions of the stomach...... 15 30 Figure 3: Small intestine...... 18 31 Figure 4: Main parameters of small intestine...... 20 32 Figure 5: Large intestine...... 21 33 Figure 6: Liver, pancreas and gall bladder...... 25 34 Figure 7: Colon regions where the mesentery is attached to...... 27 35 Figure 8: The tissue layers of the GI tract...... 28 36 Figure 9: Mucus...... 32 37 Figure 10: Different phases of gastric digestion...... 42 38 Figure 11: Segmentation motility inside small intestine...... 43 39 Figure 12: Peristalsis movement along the lumen...... 44 40 Figure 13: The three phases of interdigestive motility pattern...... 48 41 Figure 14: The forces acting on an object moving through the GI tract...... 49 42 Figure 15: Factors affecting the tribology of an object moving through the GI tract...... 56 43

48 2 49

50 Abstract

51 The gastrointestinal (GI) tract is a complex environment comprised of the mouth,

52 esophagus, stomach, small and large intestines, rectum and anus, which all cooperate to form the

53 complete working GI system. Access to the GI using endoscopy has been augmented over the past

54 several decades by swallowable diagnostic electromechanical devices, such as pill cameras.

55 Research continues today and into the foreseeable future on new and more capable miniature

56 devices for the purposes of systemic drug delivery, therapy, tissue biopsy, microbiome sampling,

57 and a host of other novel ground-breaking applications. The purpose of this review is to provide

58 engineers in this field a comprehensive reference manual of the GI environment and its complex

59 physical, biological, and chemical characteristics so they can more quickly understand the

60 constraints and challenges associated with developing devices for the GI space. To accomplish

61 this, the work reviews and summarizes a broad spectrum of literature covering the main anatomical

62 and physiological properties of the GI tract that are pertinent to successful development and

63 operation of an electromechanical device. Each organ in the GI is discussed in this context,

64 including the main mechanisms of digestion, chemical and mechanical processes that could impact

65 devices, and GI motor behavior and resultant forces that may be experienced by objects as they

66 move through the environment of the gut.

68 Keywords: digestive system, capsule endoscospy, gastrointestinal anatomy, gastrointestinal

69 properties.

71 Introduction

72 The digestive system is made up of the gastrointestinal (GI) tract, liver, pancreas and

73 gallbladder. The GI tract is a large, hollow, tubular organ system that extends from the mouth to

74 the anus. It is a complex environment that comprises the mouth, esophagus, stomach, small and

75 large intestines. These organs have specific functions and they cooperate in order to form a

76 complete working GI tract. The coordinated contractions of muscles, along with the release

77 of hormones and enzymes facilitate the digestion of food, the absorption of nutrients, and the

78 elimination of waste so that the body can carry out its functions of metabolism, growth, and repair

79 (1).

80 The inspection of the GI tract is fundamental for the early detection and diagnosis of GI

81 diseases, of which there are many. In the last decade, miniaturized robots for gastrointestinal

82 inspection have been investigated with the aim of developing innovative, more sophisticated, and

83 minimally invasive technologies to access this part of the body. Despite the progress achieved so

84 far, the need for innovation is still present and stronger than ever due to the combination of a

85 growing disease prevalence and the harsh, difficult-to-access environment of the gut.

86 According to the National Institutes of Health (NIH), more than 34 million Americans are

87 suffering from diseases of the digestive system, 20 million of which have chronic disorders (2).

88 Digestive diseases encompass more than 40 acute, chronic, recurrent, or functional disorders. The

89 most common digestive diseases are irritable bowel syndrome (IBS), inflammable bowel disease

90 (IBD) (i.e., Crohn’s disease (CD) and Ulcerative colitis (UC)), celiac disease, diverticulosis, and

91 acid reflux. It is estimated that 8% of the U.S. population have chronic digestive diseases, 6% have

92 acute episodes of digestive diseases, and 43% have intermittent digestive disorders. Only 43% are

93 unaffected. As a group, digestive diseases account for 8%-9% of total U.S. mortality, of which

94 60% is due to malignant neoplasms and 40% due to non-malignant causes (3).

95 According to International Agency for Research on Cancer (4), cancer is the leading cause

96 of death in the 21st century. The statistics present some important data about the spreading of

97 cancer worldwide in 2018. Regarding colorectal cancer, the percentage of new cases was 6.1%

98 and the percentage of deaths was 5.8%, while rectal cancer had a 3.9% incidence of new cases and

99 3.2% of deaths. Considering the stomach, the percentages were 5.7% and 8.2% of new cases and

100 the number of deaths, respectively, while esophagus cancer counted 3.2% of new cases and 5.3%

101 of deaths in 2018. The International Agency for Research on Cancer (4) also reports statistics about

102 the estimated number of new cases (incidence rate, IR) and estimated number of deaths (mortality

103 rate, MR) of each type of gastrointestinal cancer in different countries in 2018. These are reported

104 in Table 1, where the estimated number has been rounded to the nearest one hundred for the sake

105 of clarity.

106 Table 1: Incidence and mortality of gastrointestinal cancer in 2018 (4).

North America Europe Asia Africa

IR MR IR MR IR MR IR MR

Colon 179 800 64 100 499 700 242 500 957 900 461 400 61 800 40 000

Stomach 29 300 13 400 133 300 102 200 769 700 584 400 31 100 28 700

Esophagus 22 700 18 200 53 000 45 100 444 600 397 700 28 500 27 700

107

108 Different procedures are used for examining the GI tract: capsule endoscopies are used to

109 inspect the entire GI tract because of their small size, while other types of endoscopic procedures

110 are used to inspect a particular organ. For example, gastroscopy is employed for the inspection of

111 the esophagus and stomach, colonoscopy for the colon, sigmoidoscopy for the sigmoid colon and

112 small bowel enteroscopy is used for the examination of the small intestine. The purposes of the

113 devices are typically for both diagnosis and therapy, however, some lack therapeutic capabilities

114 because of device size constraints. Despite the ubiquity of these procedures, they can be stressful

115 and painful for the patient (1,5) due largely to the construction of the endoscopes, which consist

116 of semi-rigid tubes that are pushed and twisted through the body of the patient by the physician,

117 causing discomfort as the instrument deforms the sensitive GI tract.

118 According to the Center for Disease Control and Prevention (6), only 66% of Americans

119 comply with screening guidelines, and therefore, an estimated 23 million people in the United

120 States avoid these procedures. This lack of intervention increases the risk of developing a cancer

121 (7). Introducing a less–invasive procedure could increase patient compliance to the endoscopic

122 procedure by reducing procedural discomfort (and the associated anxiety), risk of adverse events,

123 and the potential need for sedation (8,9). This has motivated many to develop new technologies to

124 replace the standard flexible endoscope (10).

125 To develop new and more sophisticated devices, capsule engineers need to understand this

126 complex environment. Therefore, the purpose of this paper is to provide a summary of the critical

127 information regarding the anatomy, histology, physiology, mechanics, and chemistry of the gut as

128 it pertains to medical device engineers. Providing a comprehensive primer of the GI tract can help

129 engineers in this field to speed the development of innovative technologies.

130 The paper will describe the characteristics of the esophagus, stomach, small and large

131 intestines – the regions of primary interest to engineers that develop capsule endoscopes and

132 similar devices. Given that capsules are typically designed to travel the GI tract, the liver, pancreas 6

133 and gallbladder are not the main focus of our research but are described briefly since their

134 enzymatic secretions help with the digestion of the food. Although, focus is placed on providing a

135 comprehensive description of the healthy GI tract, the most common ways disease impacts its

136 properties and function is also discussed briefly.

137 The paper is organized as follows. In Section 1 we briefly summarize the current non–

138 invasive technologies and we discuss their main limitations. In Section 2 and Section 3, we first

139 provide a general description of the anatomy and histology of the GI tract. In Section 4 and 5

140 respectively, we present the chemical and mechanical properties of the GI tract, and in Section 6

141 we investigate the motor behavior of the gut. In Section 7 we summarize all the forces that act on

142 an untethered device in order to understand the capsule dynamics within the GI tract. Section 8

143 considers the possible physiological alterations to the GI tract from the most common and severe

144 digestive diseases. Finally, in Section 9 we summarize the work, and we briefly discuss open

145 challenges about medical robotic devices; more detailed discussions related to each topic of the

146 paper are included throughout the article.

147

148 1. Current non–invasive endoscopic technologies 149 Conventional flexible endoscopy (e.g. colonoscopy) has been widely used to inspect the

150 entire GI tract. However, despite the widespread adoption of endoscopes, issues around their

151 invasiveness lead to limitation in their ability to diagnose and treat GI disease. Therefore, the

152 demand for new, less invasive and more sophisticated procedures has increased. In the last

153 decades, completely minimally invasive methods have become commercially available for

154 diagnosing the GI tract (11) and researchers have studied appealing non-invasive alternatives to

155 traditional diagnostic techniques. The new technologies are briefly discussed in the following,

156 highlighting the major advantage and disadvantages of each technique.

157 Virtual endoscopy is a technique based on computed tomography (CT) or on magnetic

158 resonance imaging (MRI), and is used to inspect the GI tract (10,12,13). Despite it being

159 completely noninvasive and almost comparable to standard endoscopy in terms of diagnostic yield,

160 its main drawbacks are an inability to biopsy, impossibility to deliver in-situ therapy and limited

161 accuracy, particularly with small or very flat lesions.

162 Similarly, in the last 20 years, swallowable capsule endoscopes (CEs) have been developed

163 and commercialized to facilitate minimally invasive exploration of hard–to–reach regions of the

164 GI tract. To date, the most prevalent clinically used CEs worldwide are reported in Table 2 (10,14–

165 16).

166 Table 2: GI capsules in clinical use today (10,14–16).

PillCam PillCam Endo Capso Miro Navi COLON OMOM ESO3 Capsule Cam Cam Cam 2

Small Stomach bowel Esoph Small Small Small imaging Colon imaging Purpose ageal bowel bowel bowel and imaging and imaging imaging imaging imaging naviga naviga tion tion

Dimension 26.0 × 31.5 × 26.0 × 31.0 × 24.0 × 27.9 × 28.0 ×

(mm) 11.0 11.6 11.0 11.3 11.0 13.0 12.0

Weight (g) 3.4 2.9 3.8 4.0 3.4 6.0 5.0

4 high

resolu Adaptive Adaptive tion Frame Frame camera: Rate Rate maxi (AFR): (AFR): Frame rate 2 fps mum of 3 fps 0.5–2 fps 2 fps 2–6 2–6 20 frame images images per per per second (5 second. second. fps per

camera)

Angle of 160° 160° 160° 360° 170° 140° 140° view

2 x 6 2 x 6 6 white 6 white 6 white Illumination white white LEDs LED LEDs LEDs LEDs LEDs LEDs

Recording 30 min 10 h 9 h 15 h 11 h 7–9 h 8 h time

Regulatory Yes Yes Yes No Yes Yes Yes approval

167

168 Despite encouraging results obtained by current CE technologies, their main limitation lies

169 in the fact that their sensitivity (the proportion of positive cases correctly identified) is not yet

170 comparable to the sensitivity of the standard endoscopes. For example, the sensitivity of the

171 PillCam, the most advanced capsule developed, is still less than 90%, while standard endoscopes

172 can reach in excess of 95% sensitivity (10,17). Other important limitations are (10,14):

173 • Passive locomotion: the physician is not able to control the pose of the capsule (or

174 orientation of the camera) which leads to inadequate inspection of the organ in some

175 cases.

176 • Minor interventions (e.g. Biopsy collection) are the main advantage of standard

177 endoscopes and are not currently possible with commercial capsule endoscopes.

178 This is largely due to a lack of device position control, limited on-board space (or

179 payload), and the absence of a stable platform.

180 • No means of insufflation: the inability to distend collapsed tissue may lead to

181 reduced visibility, particularly in the cavernous environments of the stomach and

182 colon.

183 Several solutions have been developed for enabling active locomotion capsules and thus

184 overcome the main limitations of passive locomotion. Fundamentally, two major solutions have

185 been exploited to address the active locomotion problem: onboard locomotion (generally this is

186 referred as a mechanical approach) and an external locomotion technique (whereby forces and

187 torques are transmitted to the capsule from outside the body, generally via magnetic fields) (10,18–

188 20).

189 In the last decades, a new category of flexible endoscopes has been explored by researchers.

190 These advanced flexible endoscopes, or soft-tethered capsules, are designed to preserve the major

191 functionalities of conventional endoscopy that are familiar to physicians. At the same time, the

192 flexibility of the endoscope body permits it to conform to the shape of the bowel, reducing tissue

193 stretching and the associated discomfort for the patient (21,22).

194 Advanced flexible endoscopes have the advantage of overcoming the main limitations

195 related to the CEs (23–25). The advantage of an actuation mechanism, such as the magnetic field,

196 is the ability to pull and steer the endoscope inside the body and so completely inspect the organ.

197 Moreover, the tether (with cables and lumen) allows the physician to use the endoscope both as a

198 diagnostic or therapeutic instrument, and with all the typical auxiliary functions such as

199 insufflation, irrigation and suction. However, the presence of the tether (although soft and flexible)

200 creates friction in the environment and makes locomotion challenging (5). Other research has

201 addressed the problem of drag on the soft tether by modifying the locomotion strategy (26).

202 Aside from standard diagnostic routine, capsule robots are being used as a platform for

203 versatile applications such as drug delivery, biosensing, and active diagnostics and intervention

204 (27). Researchers have measured pH, core body temperature, oxygenation, electric

205 conductivity and, also, blood inside the intestine via capsule robots. These have extended the

206 boundary towards intervention and therapeutic manipulation (28). Clip deployment for stopping

207 bleeding, systemic and topical drug delivery, biopsy tissue collection and microultrasound imaging

208 are some other applications being investigate. Even though interventional capsules are mostly at

209 the proof-of-concept stage of development, in-vivo animal studies, and benchtop experiment

210 results are encouraging (29–32).

211 Despite the advance in current technologies the need for novel designs and innovative

212 strategies to enhance the physician’s ability to diagnose and treat GI diseases is still present.

213 Miniaturized, capsule-like devices with advanced locomotion techniques and sensing technologies

214 have been at the forefront in achieving this. For this reason, an adequate knowledge of the entire

215 GI tract is mandatory for all the engineers in the medical and endoscopic field.

216

217 2. Anatomy and physiology of the GI

218 The human GI tract is a series of multilayered, tubular organs which extend from the oral

219 cavity through the esophagus, stomach, small and large intestines and terminating at the anus

220 (Figure 1). The GI tract has a total average length of 795 ± 128 cm, it decreases with age and is

221 significantly longer in men (33). The GI tract is one of the most dynamic organ systems (34);

222 muscular contractions, along with the release of hormones and enzymes, enable the digestion

223 process (35,36). In this section, the anatomy of the GI tract will be discussed, focusing on the

224 esophagus, stomach, small and large intestine and the mesentery. The liver, gallbladder and

225 pancreas, are not the focus of this work but will be discussed briefly for the sake of clarity.

226

227 Figure 1: Anatomy of the digestive system, showing all organs.

228 2.1 Esophagus

229 The esophagus is an 18 to 25 cm long muscular tube that connects the oral cavity to the

230 stomach (36–38). It is a dynamic tube which serves to propel food, via active peristaltic

231 contractions, toward the stomach for continued digestion and absorption of nutrients. When empty,

232 the esophagus is a collapsed lumen, but it can distend to approximately 2 – 3 cm to propel a food

233 bolus (38,39) .

234 In general, the ease of passage of a body through the esophagus and into the stomach

235 depends on the length and diameter of the ingested object. Bodies longer than 60 mm and with a

236 diameter more than 25 mm make the passage difficult and objects can become lodged; in these

237 cases an esophagogastroscopy is necessary (40–42). However, the current swallowable capsules

238 approved by the FDA present smaller values for diameter and length and give a more ideal target

239 size. These values are presented in Table 2.

240 The passage of food through the esophagus is regulated by two principal high-pressure

241 valves: the upper and the lower sphincter. These two valves are located at the beginning and at the

242 end of the esophagus, but there is not a clear anatomic demarcation that defines the two zones (39).

243 The upper esophageal sphincter controls the movement of food from the pharynx into the

244 esophagus, while the lower esophageal sphincter (also called gastroesophageal or cardiac

245 sphincter) lets food pass into the stomach; the latter can also contract to prevent stomach acids

246 from backing up into the esophagus (39).

247 The normal esophagus has a wall thickness of 4.7 mm (range 4.44 – 4.95 mm) during

248 contraction and 2.11 mm (range 2.00 – 2.23 mm) when the esophagus is dilated. The thickness of

249 the esophageal wall depends also on the sex and age of the patient (43,44).

250 2.2 Stomach 251 The stomach is a dilated and J-shaped organ that rests on the left of the central region of

252 the abdomen at the level of the first lumbar vertebra (35). The main functions of the stomach are

253 the temporary storage, mixing, breakdown, and digestion of food (45).

254 As shown in Figure 2, the stomach has two openings (esophageal and the duodenal) and

255 five major regions, including: the cardia, fundus, body, antrum and the pylorus (38). The cardia is

256 the point where the esophagus connects to the stomach. The fundus is dome shaped and locates

257 inferior to the diaphragm, above and to the left of the cardia. Below the fundus is the ‘body’, the

258 main part of the stomach. The pylorus is a funnel-shaped valve which connects the stomach to the 14

259 duodenum (45). The pylorus has two parts: the pyloric antrum, which is connected to the body of

260 the stomach, and the pyloric canal, which is connected to the duodenum. The smooth

261 muscle pyloric sphincter is located at this latter point of connection and controls stomach

262 emptying. The pyloric diameter is controlled by the contractions of the sphincter and this

263 determines the flow resistance (46).

264

265 Figure 2: The functional regions of the stomach.

266 Each region of the stomach has a different function: the fundus can relax to accept large

267 volumes of collected digestive gases; the gastric chief cell in the stomach secretes pepsinogen and

268 hydrochloric acid, produced by gastric parietal cells, to break-down and mix the food and liquid;

269 the pylorus is responsible for mucus, protein-digesting enzyme (pepsin) secretion, and handles the

270 emptying of the stomach through the duodenum (45).

271 Stomach emptying is an essential factor that capsule engineers should consider. Although

272 the stomach volume is only 0.8 L when empty, it can expand up to 1.5 – 2 L for a typical male and

273 up to 0.9 – 1.5 L for women and children. The emptying rate is affected by meal composition and

274 consistency (47), body position (48), smoking (49) and gender (50). On average, women have a

275 slower gastric emptying compared to men (74 minutes vs. 63 minutes) and smokers have a

276 significantly faster gastric emptying compared to non-smokers (56 minutes vs. 67 minutes). Age,

277 body mass and alcohol consumption habits are not known to affect gastric emptying times (51).

278 The mean thickness of the gastric wall was measured as 4 mm when distended and 5 – 10

279 mm during fasting (52). In studies done by Huh et al. (53) endoscopic ultrasound was used to

280 acquire in–vivo data on wall thickness of the stomach when it was waterfilled (i.e., distended). The

281 measurements were taken in a group of ten and five measurements in different locations were

282 obtained resulting in a mean thickness of 3.92 ± 0.16 mm.

283 2.3 Small intestine 284 The small intestine (or small bowel) is a crucial component of the digestive system and is

285 responsible for the absorption of important nutrients (38). Here, most of the chemical and

286 mechanical digestion is carried out (54). It is a long, approximately 6 m, continuous, and highly

287 tortuous tube running from the pylorus of the stomach to the ileocecal valve where it meets the

288 large intestine. There are three main sections to the small intestine: duodenum, jejunum and ileum

289 (38). These sections, described below, are shown in Figure 3.

290 The duodenum is the first section of the small intestine and forms a 'C' shape around the

291 head of the pancreas. Its main function is to neutralize the acidic gastric contents (called 'chyme')

292 and to initiate further digestion (55). Brunner's glands in the submucosa secrete an alkaline mucus

293 which neutralizes the chyme and protects the surface of the duodenum. It is about 25 cm in length 16

294 in adults, beginning at the pylorus and ending at the ligament of Treitz, which is the junction

295 between the duodenum and jejunum (duodenojejunal flexure) (55,56) (Figure 3). The duodenum

296 is largely retroperitoneal and has an anatomic relationship with the pancreas. It has four sections:

297 bulb, descending, transverse, and ascending. The bulb section begins at the pylorus, which is

298 approximately 5 cm in length for adults and demarcated by the pre-pyloric vein. The descending

299 section is retroperitoneal and is approximately 10 cm in length. The transverse section is also

300 retroperitoneal and is bordered by the pancreas superiorly and the hepatic flexure of

301 the colon anteriorly. The fourth portion of the duodenum courses in a cephalad direction to the left

302 of the aorta and inferior to the neck of the pancreas. The duodenum contains a slender band of

303 skeletal muscle and a fibromuscular band of smooth muscle in the horizontal and ascending parts.

304 These can contract to widen the angle of the duodenojejunal flexure and allow movement of

305 intestinal contents (56).

306 The jejunum and ileum lie within the peritoneal cavity and are the most tortuous parts of

307 the small intestine. Together, they are approximately 4 – 6 meters long comprising approximately

308 40% jejunum and 60% ileum (55) with no clear junction between the two sections. Generally, the

309 jejunum has a thicker mucosal lining (i.e., thicker wall), larger diameters, redder color, and less

310 fatty mesentery than the ileum. Moreover, the mesentery of the jejunum is attached to the left of

311 the aorta while the mesentery of the ileum is attached to the right (57). The mucosa of these

312 sections is highly folded. These folds, called plicae circulares, slow the passage of the partly

313 digested food and increase the surface area (by 1.6 times) to aid absorption of nutrients (37). The

314 majority of plicae extends transversely around the small intestine for about 50% – 65% of its

315 circumference while some of these form complete circles and others are spiral. The largest folds

316 are about 1 cm in depth at their broadest part and usually the large and small folds alternate with

317 each other (57).

318

319 Figure 3: Small intestine.

320 The total length of the small intestine varies with age (58). Mean length at 1 year is 3.8 m,

321 at 5 years is 4.5 m, at 10 years is 5 m, and at 20 years is 5.75 m (37,58,59). It becomes longer when

322 the bowel is empty and after death; thus, use of cadaveric tissue for capsule development should

323 consider this fact. It is approximately 15 mm in diameter after 35 weeks of gestational age (60)

324 and 25 mm in adults (37,61).

325 The mean values of small intestine parameters are outlined in Table 3 (61). There is no

326 statistical difference in these bowel parameters over an age range of 17 – 73 or between men and

327 women, while some pathological effects can cause changes in these parameters (61).

328

329 18

330

331 Table 3 : Mean and standard deviation values for the small intestine parameters (diameter, wall

332 thickness, fold number per 2.5 cm, fold thickness) (37,61).

Wall Fold Interfold Diameter Fold number thickness thickness distance (mm) per 2.5 cm (mm) (mm) (mm)

Duodenum 24.8 ± 4.5 1.5 ± 0.6 4.5 ± 0.7 2.1 ± 0.6 4.7 ± 1.54

Jejunum 24.5 ± 4.2 1.5 ± 0.5 4.6 ± 0.8 2.2 ± 0.7 4.59 ± 3.56

Proximal 19.5 ± 3.6 1.6 ± 0.4 1.8 ± 0.6 1.9 ± 0.5 16.8 ± 6.75 ileum

Distal ileum 18.9 ± 4.2 1.4 ± 0.5 1.6 ± 0.5 1.8 ± 0.5 18.5 ± 7.18

Terminal 18.7 ± 3.6 1.5 ± 0.4 1.5 ± 0.6 1.8 ± 0.4 17.91 ± 7.86 ileum

333

334 As shown in Table 3, the duodenum and jejunum have similar bowel diameter, wall

335 thickness, fold number, and fold thickness. The interfold distance gradually decreases in size to its

336 smallest measurements in the terminal ileum. The bowel diameter, wall thickness, interfold

337 distance and fold thickness of the proximal ileum, distal ileum and terminal ileum are similar. The

338 parameters used in Table 3 are illustrated in Figure 4.

339

340 Figure 4: Main parameters of small intestine.

341 2.4 Large intestine 342 The large intestine (colon or large bowel) is the last part of the GI tract and, like the small

343 intestine, is tubular and tortuous in shape. It is shorter at approximately 1.5 m, has more

344 pronounced folds and a larger diameter. By the time digestive products reach the large intestine,

345 almost all the nutritionally useful products have been removed; therefore, it does not play a major

346 role in absorption of nutrients. Instead, the main purpose of the large intestine is the absorption of

347 water (approximately 1.5 L of water arrive in the colon each day), and other minerals, and + 348 the collection and excretion of waste (stool) via the anus (38,62). 푁푎

349 The large intestine has six sections, as shown in Figure 5: the cecum, the ascending colon,

350 the transverse colon, the descending colon, the sigmoid colon, and the rectum. The first and middle

351 parts of the colon are called the proximal colon. This includes the cecum, the ascending colon, and

352 the transverse colon. The last part of the colon is called the distal colon and includes the descending

353 colon, the sigmoid colon, rectum and anus. The ascending colon, descending colon, and rectum

354 are retroperitoneal and fixed in location while the other two sections are intraperitoneal and

355 therefore mobile (38,62).

356

357 Figure 5: Large intestine.

358 Eickhoff et al. (63) used a CT colonography in order to obtain more information about the

359 number of colonic flexures (defined as an acute angle < 90°), and degree of tortuosity (judged on

360 a 10–point visual analogue scale (VAS)). They found that the average number of flexures in a

361 colon is 9.6 ± 2.4, and the VAS was found to be 3.7 ± 1.9. Cases with a major number of flexures

362 and an increased degree of tortuosity are difficult to access and increase the chance of incomplete

363 colonoscopy. Moreover, Alazmani et al. (64) demonstrated that the tortuosity of the colon when

364 the patient is in supine position is higher than in the prone position.

365 Table 4 reports the diameter of the large intestine described by Sadahiro et al. (65), focusing

366 on the difference between male and female. The diameters of the descending colon, sigmoid colon,

367 and rectum are larger in males than in females.

368 Table 4: Diameter (cm) of distended large intestine (37,65)

Male Female

Cecum 4.7 ± 0.9 4.8 ± 0.8

Ascendant colon 4.8 ± 1.2 5.0 ± 2.0

Transverse colon 4.2 ± 1.2 4.2 ± 0.7

Descendant colon 3.4 ± 1.2 3.2 ± 0.6

Colon sigmoid 3.4 ± 0.6 3.2 ± 0.6

Rectum 4.0 ± 1.0 3.5 ± 1.0

369

370 Table 5 and Table 6 report the diameter and length of the large intestine, considering the

371 supine and prone position of the patient. The research by Alazmani et al. (64) demonstrates that

372 the diameter is governed largely by intra-abdominal compression and pelvic motion. Therefore,

373 changing the position of the patient from prone to supine affects the position of the internal organs,

374 and thus, the compression of the colon (64).

375 Table 5: Comparison of the colon diameter (cm) in supine and prone positions (64).

Supine Prone

Rectum 3.6 ± 0.8 3.7 ± 0.7

Sigmoid 2.6 ± 0.4 2.6 ± 0.3

Descending colon 3.3 ± 0.6 3.2 ± 0.5

Transverse colon 3.7 ± 0.4 3.6 ± 0.5

Ascending colon 4.5 ± 0.7 4.3 ± 0.7

Cecum 4.4 ± 0.7 3.8 ± 0.6

Proximal colon 4.2 ± 0.4 3.9 ± 0.5

Distal colon 3.1 ± 0.5 3.1 ± 0.4

Total colon 4.7 ± 0.5 3.5 ± 0.4

376

377 Table 6: Comparison of the colon length (cm) in supine and prone positions (37,64).

Supine Prone

Rectum 23.4 ± 6.7 23.1 ± 3.9

Sigmoid 50.6 ± 13.9 49.9 ± 11.7

Descending colon 24.2 ± 7.8 26.0 ± 7.8

Transverse colon 57.2 ± 9.3 57.3 ± 10.9

Ascending colon 21.7 ± 4.2 19.7 ± 4.0

Cecum 7.8 ± 2.9 6.9 ± 2.3

Proximal colon 86.6 ± 9.7 84.0 ± 10.2

Distal colon 98.3 ± 14.7 99.0 ± 11.8

Total colon 185.0 ± 18.3 183.0 ± 16.9

378 The mean values of the length of the entire intestine are summarized in Table 7. The study

379 by Hounnou et al. (33) shows the length of the whole intestine is longer in men than women and

380 the length decreases with age, and increases with weight while it does not vary with height.

381 Table 7: Mean and standard deviation values of the length (cm) of the intestine [17], [21].

Men Women

Avg ± std Min Max Avg ± std Min Max

Duodenum (1) 27.8 ± 6.8 17 56 25.2 ± 5.4 17 48

Jejunum-ileum (2) 643.9 ± 110.8 365 1 000 573.8 ± 97.1 280 840

Small Intestine (1+2) 670.7 ± 113.1 390 1 030 599.2 ± 98.2 298 860

Right Colon (3) 74.1 ± 17.4 40 146 71.9 ± 16.5 40 125

Left Colon (4) 94.2 ± 27.2 33 220 82.9 ± 20.1 34 123

Colon (3+4) 166 ± 36.2 80 313 155 ± 28.6 80 214

Whole Intestine (1+2+3+4) 836.7 ± 132.1 550 1 316 754.2 ± 111 378 1 013

382 Regarding the thickness of the colon wall, the studies conducted by Wiesner et al. (66)

383 report a correlation between wall thickness and colonic distention. A normal wall thickness is

384 ranged between 0.2 – 2.5 mm if the colon is distended and up to 6 mm if the colon is contracted.

385 2.5 Liver, pancreas and gall bladder

386 One of the primary functions of the liver, pancreas, and gall bladder is to assist the GI tract

387 in breaking down food into its component nutrients by secreting enzymes. These organs are

388 illustrated in Figure 6.

389

390 Figure 6: Liver, pancreas and gall bladder. 391 The liver is situated in the right upper quadrant of the abdomen and is divided into two

392 primary lobes: a large right lobe and a smaller left lobe. The liver has an important role in digestion;

393 it produces bile, a thick fluid which contains enzymes that help to dissolve fat in the intestines, and

394 metabolizes nutrients that are absorbed by the intestines (67).

395 The gall bladder is a hollow, pear shaped, 8 – 10 cm long organ that is posterior to the liver.

396 It is composed of three sections: fundus, body, and neck. Its main function is storage of bile which

397 is then released via the cystic duct, a 1 – 2 cm long canal, into the biliary duct system linked to the

398 duodenum (68).

399 The pancreas is a lobular organ that lies posterior to the stomach. It secretes fluid rich in

400 carbohydrates and inactive enzymes which become active once they reach the duodenum. The

401 hormone secretion is triggered by the duodenum in the presence of chyme. The fluids secreted by

402 the pancreas are then released into the duodenum via the pancreatic duct. The particular enzymes

403 produced by the pancreas are discussed in Section 4 (69).

404 2.6 Mesentery

405 The mesentery is a continuous set of ruffled and folded tissues that extends from the base

406 of the stomach down to the rectum. It suspends the intestines from the abdominal wall in multiple

407 regions (70). Its main functions are to fix all abdominal digestive organs, connect them to the other

408 systems and to store fat. It also helps the lymphatic system to transport lymph (fluid containing

409 white blood cells) throughout the body (71).

410 Knowing how the mesentery is organized and attached to the abdominal wall is necessary

411 in order to understand how it may impact the mechanical properties of the bowel and to quantify

412 the sensitivity of each region of the GI tract. In the study of White et al. (72) the failure stress

413 values for the mesentery in porcine models was characterized.

414 Most of the small intestine is not attached to the abdominal wall and so is mobile; however,

415 the large intestine is more fixed (where the anchorage system is deficient, the organ is mobile and

416 prone to twisting around the attached region of the mesentery). The right mesocolon is the

417 continuation of the small intestinal mesentery. The transverse mesocolon starts at the hepatic

418 flexure and this continues as the left mesocolon at the splenic flexure. The right and left mesocolon

419 are similar and both are attached to the posterior abdominal wall. The mesosigmoid comprises two

420 regions: the medial region is attached to the posterial abdominal wall while the lateral region is

421 mobile. The mesorectum terminates proximal to the pelvic floor (71). The main regions of the

422 large intestine where the mesentery is attached to are shown in Figure 7.

423

424 Figure 7: Colon regions where the mesentery is attached to.

425 3. Histology of the GI

426 Histology is the study of the microanatomy of tissues. Knowing the composition of tissue

427 is necessary in order to understand how its characteristics can affect the function of a device (e.g.

428 the surface texture) and how specific cells (or regions) can be targeted for drug delivery. In this

429 section, the histological properties of the tissues of the GI tract will be summarized. The most

430 common layers of a digestive tissue are shown graphically in Figure 8; however, subtle differences

431 between organs are elaborated on in subsequent paragraphs.

432

433 Figure 8: The tissue layers of the GI tract. 434 3.1 Esophagus

435 The wall of the esophagus is composed of four layers: mucosa, submucosa, muscularis

436 propria, and adventitia. With respect to the other organs of the GI tract, the esophagus is the only

437 one that does not have a serosa layer. The missing serosa layer allows esophageal cancer to spread

438 easily and for this reason the surgical treatment is more challenging. Without a serosa layer,

439 possible luminal disruptions are also more difficult to repair.

440 The mucosa is thick and red at the beginning and paler at the end of the esophagus.

441 Longitudinal folds are present in the mucosa but they disappear when the esophagus is distended.

442 The mucosa consists of three sublayers: mucous membrane, lamina propria and muscularis

443 mucosa. The submucosa is made up of connective tissue, cells such as lymphocytes and plasma,

444 and mucous glands. The secretion of these glands is important for the clearance of the esophagus

445 and for tissue resistance to acid. The muscularis propria, which is responsible for motor function,

446 is composed of striated (skeletal) muscle in the upper part and of smooth muscle in the lower part

447 of the esophagus. The middle area, called the transition zone, is a mixture of both muscles. The

448 adventitia is the external fibrous layer and connects the esophagus with the surrounding

449 environment. Therefore, it is composed of connective tissue, small vessels, lymphatic channels,

450 and nerve fibers (39).

451 3.2 Stomach

452 The wall of the stomach consists of four layers: mucosa, submucosa, muscularis propria

453 (or muscularis externa), and serosa (73). The mucosa is relatively thick and contains numerous

454 gastric glands and pits. The mucosa of the stomach has a mean thickness of 1.26 ± 0.07 mm and

455 accounts for about 32% ± 7% of the total thickness of the stomach (53). It has a prominent layer

456 of smooth muscle called muscularis mucosa, which helps to expel the contents of the gastric

457 glands. The mean thickness of the muscularis mucosa is 0.17 ± 0.09 mm (74). The submucosa,

458 made up of connective tissue and lymph vessels, separates the mucosa from the muscularis externa.

459 The muscularis externa consists of 3 layers of smooth muscle: inner oblique layer, middle circular

460 layer, and external longitudinal layer. The three layers are not always visible but have different

461 functions: the inner oblique layer helps to mechanically break down food; the middle circular layer

462 of the muscularis is thick and forms the pyloric sphincter; and the external longitudinal layer is

463 responsible for moving the bolus towards the pylorus of the stomach (74). The serosa is the

464 outermost layer that covers all the stomach wall (73).

465 3.3 Small intestine

466 The small intestine also has four tissue layers: mucosa, submucosa, muscularis propria, and

467 serosa, as shown in Figure 8. The mucosa secretes digestive enzymes and hormones, and has many

468 protrusions called villi. These dramatically increase the surface area of the small intestine (by 60

469 – 120 times) helping the absorption of the digested food (37). This layer is the thickest and can

470 make up 35% – 40% of the overall wall of the small intestine. The submucosa is the layer of dense,

471 irregular connective tissue or loose connective tissue and contains mucous glands, blood vessels,

472 lymph vessels, and nerves. It supports the mucosa and joins the mucosa to the underlying smooth

473 muscle. The muscularis propria is a region of muscle nearby the submucosa membrane. It usually

474 has two distinct layers of smooth muscle (circular and longitudinal) and is responsible for

475 peristaltic movement. The serosa is the outermost layer of the intestine: it is a smooth membrane

476 consisting of a thin layer of connective tissue and a thin layer of cells that secrete serous fluid (73).

477 The three main sections of small intestine are similar at a microscopic level. Therefore, it

478 can be assumed that the previously mentioned sections of the small intestine have layers of the

479 same thickness. The mucosa and the submucosa have a thickness of 0.4 ± 0.1 mm while the

480 thickness of the muscularis propria is 0.4 ± 0.2 mm (75). The three sections have slightly different

481 functions. For example, unlike in the jejunum and ileum, the submucosa in the duodenum has

482 Brunner’s glands whose main function is to produce a mucus-rich, alkaline secretion to neutralize

483 the acidic content of chyme introduced from the stomach and to provide an alkaline condition for

484 optimal intestinal enzyme activity for enabling absorption. On-the-other-hand, the ileum has

485 Peyer’s patches in the mucosa whose function is the immune surveillance system of the intestinal

486 lumen (73).

487 3.4 Large intestine

488 The histology of the large intestine is similar to that of the small intestine. However, since

489 the function of the large intestine is to absorb water there is no plicae circulares or villi. Therefore,

490 compared with the small intestine, it is more uniform and flatter on the microscopic scale (38,73).

491 The mean thickness of the large intestine wall is 1080 ± 239 µm in which the mean thickness of

492 the mucosa is 499 ± 104 µm, the thickness of the muscularis mucosa is 62 ± 32 µm, and the

493 submucosa is 519 ± 234 µm (74,76). The mucosa is composed by a thin layer of epithelial cells

494 (epithelium), connective tissue (lamina propria), and muscle (muscularis mucosa). The submucosa

495 surrounds the mucosa, and it is made up of mucous glands, blood vessels, lymph vessels, and

496 nerves. The muscularis propria is a layer of muscle that surrounds the wall of the colon and rectum.

497 The serosa is the outer layer of the colon that it is not found on most of the rectum (73).

498 3.5 Mucus

499 Mucus is an essential factor to consider in device development, having a direct impact on

500 the navigation of a device inside the GI tract and its interaction with the tissue for diagnosis and

501 treatment (e.g. drug delivery). Mucus is present on all surfaces of the GI tract and creates a physical

502 barrier between the epithelium and the object in contact with it. For navigation, this can result in a

503 slippage plane that facilitates the easy passage of the object through the GI tract, protecting the

504 tissue from mechanical wear. Alternatively, it can be utilized for the opposite – leveraging muco-

505 adhesion to gain traction for locomotion or anchoring. For diagnosis and treatment, this layer can

506 be a source of information on gut health, or a physical barrier through which the tool, sensor or

507 drug must penetrate.

508

509 Figure 9: Mucus. 510 Mucus (Figure 9) is a complex biological material and its main functions are the lubrication

511 of the tissue in order to transport the chyme from the esophagus to the colon (77), and the creation

512 of a barrier to protect the surfaces of the GI tract and control the bacterial interaction with the

513 immune system (77,78). The mucus is a semipermeable barrier that enables the exchange of

514 nutrients, water, gases, and hormones, but at the same time, it is impermeable to most bacteria and

515 pathogens (79). The mucus has an important role in drug delivery since it behaves as a barrier to

516 some molecules and thus, drugs. Its viscoelasticity and pH properties can impact the delivery and

517 absorption of drugs (80–82).

518 The composition of the mucus varies markedly along the GI tract. The mucus is normally

519 composed of water, and so, it becomes a viscous structure when dehydrated (78). The mucus is

520 composed of mucins, a protein that gives gel-like properties to the mucus. In the mouth, the

521 salivary glands produce MUC5B and MUC7, which lubricate the food in order to pass through the

522 esophagus. The stomach and colon have a two-layered system, and the major component of the

523 colon mucus is the MUC2 mucin while the MUC5AC mucin is the major component of the

524 stomach mucus. Both are produced by the goblet cells (78). The small intestine has, in contrast to

525 the stomach and colon, only one type of surface mucus, composed of MUC2 (83).

526 The two-layered structure of mucus in the colon is noteworthy (78): the outer layer is

527 permeable, and therefore, is the typical habitat for bacteria; however, the bacteria in the colon do

528 not have any direct contact with the epithelial cells since the inner mucus layer is impermeable.

529 The inner mucus layer is continuously secreted from the goblet cells. The inner layer of both the

530 colonic and stomach mucus is attached to the epithelial cells and is not easily removed, while the

531 outer layer easily sloughs off. The outer layer of the colon is easier to remove compared to the

532 outer layer of the stomach. The mucus of the small intestine does not normally adhere to the

533 epithelial cells and it is easier to remove (78). The mucus in the small intestine covers the overall

534 space between the villi, and since it is not anchored to the epithelial cells, it moves with the

535 peristaltic waves. However, new mucus is constantly produced from the goblet cells. Here, the

536 mucus is also formed by antibacterial proteins whose function is to limit the number of intact

537 bacteria that can reach the epithelium (78).

538 In humans the thickness of the colonic inner layer is about 200 – 300 µm (78). The

539 spontaneous mucus growth is 240 ± 60 µm/h and the final mucus thickness is 480 ± 70 µm in the

540 colon (84). The mucus of the stomach has a mean value of 180 µm with a range of 50 – 450 µm.

541 The thickness depends mainly on digestive activity in the small intestine (79).

542 The viscoelasticity of the mucus depends on the level of hydration and on mucin concentration

543 (79). The slope of viscosity versus the shear rate for mucus is usually within the range of -1 to -

544 0.5, with an average of -0.85. The viscosity of healthy gastric mucus is about 0.085 Pa s at a shear

545 rate of 1.15 , but this value can increase significantly during duodenal ulceration (77). −1 33 푠

546

547 4. Chemical makeup of the contents of each region

548 The chemical properties of the GI tract, such as the pH, the enzymatic composition and the

549 metabolic activity, are crucial for determining appropriate materials for the design of the device,

550 selecting sensors and for choosing a location inside the gut for targeted drug delivery.

551 4.1 pH

552 The pH has a crucial role in the digestive tract, helping to create a favorable environment

553 for the breakdown of food and controlling bacteria metabolism. The pH along a healthy gut is

554 presented in Table 8 (85–87). The saliva has a near neutral pH, but the oral cavity pH may be

555 modified by food. Secretion of different enzymes and chemicals controls the overall pH profile of

556 the gut. Regarding the esophagus, the normal value of pH is between 6.0 and 7.0 but it can drop

557 down to 4.0 in the presence of gastroesophageal reflux (88,89).

558 Table 8: The pH values at different locations of the human gut.

Location pH (mean ± SD)

Stomach (85) 2.9 ± 1.97

Duodenum (86) 6.6 ± 0.5

Jejunum (85) 7.1 ± 0.6

Ileum (fold) (86) 7.5 ± 0.4

Large intestine (86) 6.6 ± 0.7

559

560 4.2 Chemicals and Enzymes

561 Digestion is a complex process and consists of both mechanical and chemical mechanisms.

562 The former is relatively simple and involves physical breakdown of food through muscular

563 contractions. The latter is a more complex mechanism that reduces food into its chemical

564 components which are then absorbed. In healthy individuals a substantial amount of fluid and ions,

565 about 7 L, is secreted and reabsorbed daily by the GI tract.

566 Chemical digestion begins first in the mouth by means of the salivary enzyme amylase

567 which breaks down starches into glucose. The esophagus does not produce digestive enzymes but

568 does produce mucus for lubrication and protection as food travels to the stomach (90).

569 The cells in the lining of the stomach wall secrete hydrochloric acid (HCl), potassium

570 chloride (KCl), and sodium chloride (NaCl). Combined, these are known as gastric acid.

571 Bicarbonate, a base, is located to buffer the gastric fluid and mucus, a viscous fluid, protects the

572 stomach wall. The gastric chief cells in the stomach release pepsinogen and gastric lipase that help

573 to digest protein and lipid, respectively. Also, amylase, produced in the oral cavity and transferred

574 to the stomach with food, helps to continue the digestion of starch. A healthy adult human secretes

575 about 1.5 L of gastric fluids per day (90).

576 The intestinal gland, placed between the villi of the small intestine, secretes a solution

577 almost similar to interstitial fluid. The villi contain goblet cells that produce mucus. Intestinal

578 epithelium produces various enzymes (i.e., enterokinase, disaccharidases, and peptidases). Daily

579 volume of total intestinal secretion is about 1.8 L. These enzymes are mostly secluded within the

580 cells and do not contribute to luminal flow. The exocrine enzymes produced in the pancreas, along

581 with sodium bicarbonate, are propelled into the duodenum.

582 The pancreatic enzymes consist of amylase, lipase, colipase and phospholipase, cholesterol

583 esterase, trypsinogen, chymotrypsinogen, and carboxypolypeptidase. A total of 1.0 – 1.5 L of fluid

584 are secreted each day. Also, about 1.5 L of bile are secreted every day in the liver and the excess

585 is stored in the gall bladder. Bile flows to the small intestine in the presence of fats in the

586 duodenum. Bile contains water, bile salts, bile pigments, cholesterol, inorganic salts, fatty acids,

587 fat, and lecithin (90).

588 The large intestine secretes about 0.2 L of fluid per day, mostly in form of mucus, as the

589 primary function is the absorption. It can absorbs a large amount of water, electrolytes and minerals

590 secreted from other regions, but no chemical digestion is carried on in the large intestine (91).

591 A summary of the key enzymes and chemicals are presented in Table 9.

592 Table 9: Summary of enzymes and chemical composition at different locations of the human gut

593 (91).

Location Daily volume (cc) Enzymes Fluids and ions

Hydrochloric acid

Pepsinogen Potassium chloride Stomach 1 500 Gastric Lipase Sodium chloride

Mucus

Enterokinase Mucus Small Intestine 1 800 Disaccharidases Intestinal fluid Peptidases

Large intestine 200 - Mucus

Pancreas 1 500 Amylase Sodium bicarbonate

Lipase, Nucleases

Cholesterol esterase

Tripsinogen

Chymotripsinogen

Carboxypolypeptidase

lactate dehydrogenase

Liver 1 500 aspartate and alanine Bile

aminotransferases

594

595 4.3 Gut microbiota and metabolites

596 Gut microbiota play a major role in human physiology by producing vitamins, facilitating

597 digestion, modulating the mucosal immune system and contributing to host defense against

598 pathogens (92,93). A healthy human gut hosts trillions of microbes which are essential for

599 maintaining immune and metabolic homeostasis and protecting against pathogens (94).

600 The esophagus is an environment that contains a consistent quantity of microbiota. The

601 major component of the microbiota in a healthy esophagus is Streptococcus (95). The human

602 stomach has acidic conditions and other antimicrobial factors and has been viewed as an

603 inhospitable environment for microorganisms. However, a diverse community, as large as 128

604 phylotypes among eight bacterial phyla, have been detected in the human stomach, such as

605 Proteobacteria, Firmicutes, Actinobacteria, Bacteroidetes, and Fusobacteria phyla (96). The small

606 intestine microbiota contains a facultative and strict anaerobes mainly consisting of Streptococcus

607 sp., Escherichia coli, Clostridium sp., and high G+C organisms (97). These microbes have

608 developed different survival strategies to survive the harsh environment of the small intestine. A

609 total of 395 bacterial phylotypes are identified in large intestinal mucosal and fecal samples,

610 consisting mainly of Firmicutes and Bacteroidetes phyla (98,99). Only a few sequences associated

611 with the Proteobacteria, Actinobacteria, Fusobacteria, and Verrucomicrobia phyla were found due

612 to the strict anaerobic mucosal regions (99).

613 The GI tract is not only a food digesting and absorbing system; it is also an endocrine organ

614 which secretes hormones in control of various metabolic processes and a lymphoid organ which

615 modulates the microbial control of host metabolism. Dietary substrates metabolized by the gut

616 microbiota comprise carbohydrates, amino acids, fatty acids, and phytochemicals. Some outputs

617 of the microbiota metabolism are acetate, propionate, butyrate derived from carbohydrates,

618 valerate, and caproate derived from amino acid (100).

619

620 5. Passive mechanical properties

621 The mechanical properties generally describe the ability of a tissue to resist deformation.

622 However, the stress-strain behavior is complex and understanding the hyperelastic nature of the

623 tissue is essential for the study of the locomotion of a device and the mechanical interaction with

624 the tissue. This can inform both the real-time control of the device and modelling during the early,

625 conceptualization stage. In this section, the passive mechanical properties of the hollow organs of

626 the GI tract are summarized.

627 5.1 Stress–strain behavior

628 The multi-layer structure of the GI tissue results in a complex stress-strain behavior that

629 not only varies with strain rate, but also depending on the region of the GI tract and direction of

630 stress applied. This is due largely to the fact that each layer of the GI tract has distinct mechanical

631 properties which allow different tissue to bear different deformation and stress (101).

632 The mucosa is loosely adherent to the underlying structures in most areas and cannot

633 withstand large stress. The submucosa has a mobile lattice of collagen fiber bundles with two

634 layers of muscle lining: circular and longitudinal. This allows the submucosa to resist significant

635 mechanical stress, but for a short duration. The serosa is typically the thinnest layer of the wall and

636 hence contributes the least to the overall tissue wall strength (102). In summary, the mechanical

637 strength of the bowel wall is determined largely by the submucosa and muscular layers while the

638 serosa and mucosa have no significant strength (103). In Table 10, the values of maximal stress

639 and destructive strain are provided for longitudinal and circumferential specimens of different

640 locations of the gut. Herein, the values of longitudinal and circumferential testing of surgically

641 removed stomach specimens are practically identical. On the other hand, stress and strain

642 characteristics for small and large intestines, and the esophagus, vary significantly depending on

643 direction of the load (i.e., are anisotropic) (101).

644 Table 10: Maximal stress and destructive strain for different locations of the gut.

Location Maximum ultimate stress (MPa) Ultimate strain (%)

Esophagus (cervical part) 2.19 ± 0.06 (Longitudinal) 70.0 ± 7

(104) 1.41 ± 0.05 (Circumferential) 82.5 ± 9

0.67 ± 0.19 (Longitudinal) 93.3 ± 18.57 Stomach (101) 0.5 ± 0.12 (Circumferential) 103.12 ± 20.23

0.548 ± 0.329 (Longitudinal) 85.76 ± 18.6 Small intestine (101) 0.92 ± 0.48 (Circumferential) 84.02 ± 19.73

1.188 ± 0.302 (Longitudinal) 40.94 ± 12.5 Large intestine (101) 0.645 ± 0.165 (Circumferential) 87.85 ± 27.0

645

646 5.2 Viscoelasticity properties

647 A time- dependent mechanical test, performed on a excised porcine esophagus, showed

648 that the esophagus has quasi-linear viscoelastic properties (105). Results showed that the stress

649 relaxed by 20 – 30% of the peak within the first 10 s and stabilized at 50% of the peak after 300

650 s. In a study of porcine stomach, it was shown that a higher stress relaxation∼ rate appeared in the

651 first 100 s, and it was about 70% of the total (106). An in-vitro porcine study found that small

652 intestine tissue relaxes a lot faster than stomach or esophagus. With an increased shear strain from

653 50% to 200%, all stress curves decrease exponentially from their highest points to some steady

654 states at ~20% within two seconds (107).

655

656 6. Motor behavior of the GI

657 The muscle contractions due to peristaltic movement are described in this section. These

658 movements impact the navigation of the device inside the GI tract and must be taken into

659 consideration during design. This includes the study of appropriate materials, device dimensions,

660 device shape, and effective control strategies that ensure the device is able to cope with the

661 movement of the organs while navigating the GI tract.

662 6.1 Peristalsis and the migrating motor complex

663 Peristalsis in the GI tract comprises a series of propagating muscular contractions which

664 help with the digestion and transportation of food. Each part of the GI tract has a distinct type of

665 motility and these are described in the following.

666 The stomach can be divided into two functional regions: gastric reservoir and gastric pump.

667 The primary function of the gastric reservoir is to aid in digestion of the food (108). The reservoir

668 stores the food and then this is processed through a series of acids and enzymes secreted from the

669 gastric wall. The secretions act as a non-immunological defense against invading pathogens, and

670 food is processed for a complex diet. The primary function of the gastric pump, which is

671 anatomically provided by the antrum and the pylorus, is gastrointestinal motility, or rather the

672 transmission of the food through the intestine (108).

673 The food bolus is transferred to the distal part of the stomach with the help of tonic

674 contractions as shown in Figure 10. Tonic contractions are sustained contractions lasting from

675 several minutes to several hours. In the distal part of the stomach, peristaltic waves – muscular

676 contractions initiated by spontaneous electrical waves – are generated in order to move chime.

677 These are generated from a particular type of cell called interstitial cells of Cajal (ICC) (109,110).

678 These cells generate a potential within their membranes called the electrical pacesetter potential

679 (111). This potential drives the electrical events within the smooth muscle of the stomach and also

680 determines the frequency and velocity of the slow waves in the distal part of the stomach (112). In

681 Cheng’s study (113), a laparoscopic device had been used to record these values and found that,

682 for humans, the frequency of the waves is 2.83 ± 0.35 min-1 and the propagation velocity is 3.0 –

683 8.0 mm s-1.

684

685 Figure 10: Different phases of gastric digestion.

686 Like the stomach, the intestines also have an ICC network between the tissue layers. The

687 ICC produces electrical pacesetter potential which generates slow waves inside the intestine.

688 Additionally, there are two types of motility in the intestines – segmentation and peristalsis.

689 Segmentation is a mixing type of motility. The chyme moves back and forth through successive

690 relaxation and contraction cycles of the stomach, as shown in Figure 11. In this type of movement,

691 the inner muscle mass aids in the contraction and in the constricting of the food bolus. In the distal

692 part of the duodenum, the frequency of segmentation is approximately 12 contractions min-1, and

693 for the ileum it is 3 – 4 contractions min-1 (114), (115).

694

695 Figure 11: Segmentation motility inside small intestine.

696 Peristalsis moves the chyme from one segment of the lumen to the forward segment, as

697 shown in Figure 12; it is called progressive movement. In order to generate this type of movement

698 there is a sequential contraction and relaxation just like the segmentation motility; however, here

699 the outer muscle layer contracts and shortens, while the inner layer relaxes and widens. The motion

700 waves are generated along the entire length of the GI – from the mouth to the anus. There are two

701 types of peristaltic waves: the basic peristalsis that moves only 10 cm along the small intestine at

702 each contraction of the intestine and the “peristaltic rushes” that occur occasionally and move

703 along the entire bowel with a high amplitude. The average velocity of basic peristalsis is around 1

704 – 2 cm min-1 (20) and the peristaltic rushes are around 2 cm sec-1 (116).

705

706 Figure 12: Peristalsis movement along the lumen.

707 The chyme, after passing the ileocecal valve, reaches the large intestine. Here the motility

708 action is not as active as the stomach or small intestine. Colonic motility consists of three types of

709 contractions: the rhythmic phasic contractions (RPCs), the giant migrating contractions (GMCs)

710 and the tonic contractions. The first causes slow net distal propulsion, the second occurs

711 infrequently but produces mass movements, and the third aids RPCs in their motor function (117).

712 According to Sarna et.al. (117), the average frequency of giant migrating contractions is around 6

713 – 10 per day and each contraction lasts, on average, for 20 seconds. The propagation velocity at

714 the distal part of the colon is about 1 cm sec-1.

715 In a study by Rao et al. (118) the number of peristaltic contractions occurring in healthy

716 humans during 24 hours was considered. The frequency of contractions increased after waking

717 and a meal, while it decreased in the colon during sleep, when motor activity is reduced (118). 44

718 The frequency and the velocity of propagation in different sections of the GI tract are

719 summarized in Table 11.

720 Table 11 : Frequency and propagation velocity of different motilities in the human GI tract.

Region Motility Pattern Frequency Velocity

Stomach Tonic Contraction 2.83 ± 0.35 min-1 3.0 – 8.0 mm s-1

(113,119)

Segmentation in Duodenum 12 min-1 12 cm min-1

(115)

Small intestine Segmentation in Ileum (115) 8 min-1 -

Segmentation in Jejenum (116) - 6 cm min-1

Peristalsis (20) - 1 – 2 cm min-1

Rush peristalsis (116) 1.4 – 2.8 cm sec-1

Colon Strong peristaltic movement 6 – 10 day-1 1 cm s-1

(117,118,120)

721

722 6.2 Transit time through the various regions

723 Transit time is the time that it takes food to travel from the mouth through the digestive

724 system to the anus. This can vary greatly between individuals and depends also on the composition

725 of the meal.

726 Fryne et al. (121) measured the transit time through various regions of the GI using a

727 magnetic tracking system. The observed gastric time was 35.5 min (range 4 – 73 min) and the

728 transit time for the small intestine was 261 min (range 241 – 402 min). They also measured the 45

729 motility data of the small intestine due to peristalsis. The propagation velocity was reported as 2.2

730 cm min-1 during post-prandial state and 2.3 cm min-1 during fasting phase. In addition, they

731 measured the contraction frequency of the stomach and intestine. The measured value for the

732 stomach was 2.85 ± 0.29 min-1 and for the intestine was 9.90 ± 0.14 min-1 post-prandial and 10.53

733 ± 0.29 min-1 during fasting. In a study by Degen and Phillips (122) it was demonstrated that there

734 is not a substantial difference between the transit time in men and women. The gastrointestinal

735 emptying time, measured with different techniques, is shown in Table 12.

736 Table 12 : The gastric and intestinal emptying time

Author Device Gastric Intestinal

(min) (min)

Fryne et. Al. (121) Pillcam 57.5 275

MTS -1 56 255

Magnetic Pill 35.5 260.5

Maurer et. Al. (123) Radiolabeled meal - 231

Miller et. Al. (124) Lactulose Breath test - 234

Camilleri et. Al. (125) Resin pellets 164 168

737

738 Most of the devices that have recorded data for transit time have been used in fasting states.

739 In real-life scenarios, the diet has to be taken into consideration. Krevsky et al. (126) used a

740 different approach to measure the transit times through different sections of the GI tract. Human

741 volunteers ingested food containing indium pellets and the transit times through various sections

742 of small bowel were determined by measuring the radioactive signal. The data showed that the

743 transit time to fully empty the stomach was 120 – 180 minutes and for emptying 50% of the small

744 intestine was 150 – 180 minutes. The transit time through the colon was about 300 – 360 minutes.

745 Cummings et al. (127) performed a study in which 12 human subjects were fed, after each meal,

746 with radio-opaque pellets for several weeks of controlled diet and measured the transit time. There

747 were three different types of diet: Ad libitum diet (i.e., free-feeding or feeding on demand),

748 standard diet, and high fiber diet. The mean transit time for each diet of the 12 individuals is shown

749 in Table 13.

750 Table 13 : Mean transit time (days) calculated from marker size (127).

Ad libitum diet Standard diet High fiber diet

(studies 1-6) (studies 7-12) (studies 7-11)

Mean Range Mean Range Mean Range

2.1 0.7 – 3.1 2.8 1.9 – 3.6 2.3 2.0 – 2.7

3.1 2.3 – 4.0 2.1 1.3 – 2.6 1.8 1.5 – 2.3

2.1 1.4 – 2.7 2.1 1.2 – 2.6 1.6 1.3 – 2.0

2.1 1.2 – 2.6 2.9 2.1 – 3.7 1.7 1.3 – 2.1

2.2 1.3 – 3.5 2.5 1.9 – 3.3 1.0 0.7 – 1.6

2.4 1.7 – 3.2 3.5 2.5 – 4.8

751

752 6.3 Post-prandial and fasting states and their effect on motor behavior

753 The peristaltic motion still pertains during fasting, but it is different in action and timing

754 than during the post-prandial state. The movement is propulsive – originating from the pylorus up

755 to the ileum – and is called the migratory motor complex (MMC) (128). It is a kind of

756 “housekeeping” movement in which the MMC sweeps away any leftover food inside the lumen.

757 This is a critical activity as a stagnant bolus can cause bacterial growth inside the lumen. The MMC

758 is generated by a hormone called “motilin” which is secreted during the fasting state. This state

759 lasts over a period of 90 – 120 min (128). During fasting, the MMC occurs in repeated cycles. This

760 cyclic pattern, as shown in Figure 13, is divided into three phases. Phase I is the motor quiescent

761 period lasting 40-60% of the cycle length and when slow waves are rarely associated with spikes.

762 Phase II presents irregular contractions in the small intestine and lasts 20-30% of the cycle length.

763 Phase III is the MMC characterized by spikes and contractions and lasts for 5 – 10 min (129).

764

765 Figure 13: The three phases of interdigestive motility pattern. 766 During the fasting and post-prandial states, the proximal and distal colon experience two

767 different motor activities, namely: ‘tonic’ and ‘phasic’. The former consists of long contractions

768 lasting for several minutes up to hours; the latter comprises brief periods of relaxation and

769 contraction. During fasting, the motility is similar between the proximal and distal colon. On the

770 contrary, during the post–prandial, the distal colon experiences an increase in the phasic motor

771 activity. In addition, the tonic activity, due to the meal, causes immediate tonic contraction in the

772 proximal and distal colon (130).

773 48

774 7. Forces on objects moving through the GI

775 The motion of a device through the GI tract is strictly related to all the forces acting on it

776 in the environment, of which there are many. The forces are highly variable and often too complex

777 to predict. Although they are derived from a variety of sources, they can be interdependent. These

778 are separated here into passive forces – those that are not generated as a result of the movement of

779 the object (e.g., muscular contractions in the bowel wall) and active forces – those that are (e.g.,

780 friction). These are summarized in Figure 14.

781

782 Figure 14: The forces acting on an object moving through the GI tract. 783 784 The aim of this chapter is to summarize all these forces applied to the object and provide

785 the key factors affecting their magnitude. While it is challenging to predict them all accurately, it

786 is useful to have a broad understanding of them during the mechatronic design of a device for this

787 environment.

788 7.1 Passive forces 789 The passive forces acting on an object are shown with black arrows in Figure 14. They

790 include gravity, buoyancy, muscular contractions generated by the GI, abdominal pressure,

791 intraluminal pressure, and mucosal adhesion.

792 7.1.1 Gravity and buoyancy 793 The most constant and simple forces to define are gravitational force and buoyancy. The

794 magnitude of gravitational force is proportional to the mass of the object, and the direction is

795 always downward in the world frame. Buoyancy opposes gravity and is calculated as

796 , where V is the volume of the object and is the density of the fluid surrounding it,

797 which퐹푏 = 휌푉푔 varies slightly along the GI region. In general, the휌 fluid can be assumed to be a Newtonian

798 fluid with a density of 1Kg L-1 (131).

799 Since the directions of gravity and buoyancy are always along the vertical axis, how they

800 impact the object’s dynamics depends on the orientation of the object and the surrounding

801 environment. For example, if the lumen and object are horizontal, they can align with and impact

802 object-tissue contact forces; if the lumen and object are vertical, they align with and can impact

803 propulsive force for locomotion.

804 7.1.2 Abdominal pressure 805 The GI tract runs through the core of the body and as such, passes by other organs and soft

806 tissues, all having mass and some of which are moving. The abdominal pressure exerted on an

807 object is the summation of the mass of tissue above the object (assuming bones are self-supporting)

808 and the forces generated by muscular contractions in the environment. The former could be

809 approximated by knowing the volume and density of the tissue above the object. Densities of soft

810 tissues range from 0.95 g cm-3 to 1.05 g cm-3 (132), and volumes can be approximated by medical

811 imaging and device localization. The latter include sources such as the beating heart, contracting

812 diaphragm and skeletal muscle movements. This component is challenging to quantify, as it is

813 dependent on the individual’s physiology, level of activity during the procedure, the orientation of

814 the body, and the pose of the object within the body.

815 A simpler approximation can be made by considering the abdomen as a whole and

816 measuring the intra-abdominal pressure (IAP) – a clinical parameter that is typically measured by

817 monitoring the pressure in the bladder. In a healthy adult, the pressure ranges from 5 – 7 mmHg

818 but can vary considerably, particularly in ill and obese patients, where values can be > 25 mm Hg.

819 Body posture can also have a significant impact on IAP, especially if the individual is lying prone

820 or if the individual is inclined (or standing) (133–135). Muscular contractions can greatly alter

821 IAP, with one study showing that during coughing and forced expiring, values of 46 mm Hg and

822 36 mm Hg respectively can be seen (136).

823 7.1.3 GI muscular contractions 824 GI muscular contractions are described in Chapter 6 and are mostly prominent in the small

825 intestine. They are primarily generated by the myenteron (muscular layer of the intestine), which

826 creates pendular movements, segmental contractions, peristalsis, and gradual reflexes (137,138).

827 To estimate their effects on the object dynamics, it is necessary to understand the magnitude, shape,

828 and frequency of the contact force (139). A general theoretical model of a solid bolus transported

829 by peristalsis was formulated by Bertuzzi (139). Miftahof et al. described bolus transport models

830 specific to the GI tract to predict contact forces (137,140,141).

831 7.1.4 Intraluminal pressure 832 Gases and liquids in the GI tract can become pressurized and exert forces on the

833 surrounding tissue and object. These can be artificially generated (e.g., insufflation from an

834 endoscope) or naturally generated (e.g., as a result of chemical processes in the gut). The primary

835 impact of intraluminal pressure is a reduction in contact pressure on the object as it counteracts the

836 other surrounding contact pressures, including those mentioned above. This is an important factor

837 to consider as the net contact pressure greatly impacts the degree of object-tissue contact, tissue

838 deformation, and therefore, both adhesion and the active forces on the object.

839 7.1.5 Summary - Net contact pressure 840 This subsection gives an indication of the expected contact pressures experienced by a

841 capsule in the small intestine - the region with the highest expected contact pressure due to its

842 muscular contractions and small lumen diameter. In other words, this gives a practical example of

843 the summation of pressures described in previous section. A device called the migrating motor

844 complex force sensor (MFS) was used to measure the force per centimeter of length exerted by the

845 small bowel on a capsule-like object (142–145). The contact force depends on the position of the

846 body, and the distal small bowel exerts 92% more contact force against the capsule than the

847 proximal small bowel, with the primary reason being that the distal small bowel has a smaller

848 diameter than the proximal small bowel (143). In Table 14 the mean values of the contact force

849 measured with different techniques in different works are summarized.

850 Table 14 : The contact force on capsule.

Author Length of capsule Contact force

Calio et al. (146) 33 mm 0.25 N cm-1

Terry et al. (143) 35 mm 0.9-2.9 N cm-1

Miftahof et al. (140) 35 mm 0.15-1.9 N cm-1

851

852 7.1.6 Adhesion 853 Mucus, described in Section 3.5, lines the inner surface of the GI and is continually secreted

854 by goblet cells (78). The glycoprotein molecules in the mucus have an ability to adhere to solids

855 because of their hydrophilic and viscoelastic properties. Mucosal adhesivity is the interfacial

856 ability to bond with a solid surface. It is measured by the energy required to separate the two

857 adhered surfaces and can be affected by several factors, such as hydration, mucus surface tension,

858 wettability, temperature, and dwell time (the amount of time the mucosa is in contact with the solid

859 surface prior to separation) (147). Mucosa adhesivity can be useful in device design; for example,

860 it can be a solution to increase static friction to avoid migration of the capsule inside the GI tract.

861 The inherent adhesivity between a capsule and mucosa was investigated by changing the

862 factors of adhesive modality (peel and tack), material (polycarbonate, micropatterned

863 polydimethylsiloxane, stainless steel, and mucosa), and bowel region (proximal, middle, and

864 distal). The results show the mean tack strength of the mucosa to engineering materials was 0.198

865 ± 0.070 mJ cm-2. The mean peel strength was 0.055 ± 0.016 mJ cm-2 (148). As the results suggest,

866 the adhesive tack strength between the mucosa and other material is larger than the peel strength.

867 Mucus thickness has some influence on mucoadhesion performance which is an important

868 factor to consider given the varying thickness throughout the GI tract. Varum at al. (149)

869 performed experiments on pigs, which is the closest model to human mucosa, in order to evaluate

870 the mucoadhesion. The experimental results showed the mean detachment forces are dependent

871 on mucus thickness: 0.084 ± 0.025 N for the stomach, 0.0575 ± 0.0125 N for the small intestine

872 and 0.066 ± 0.009 N for the colon (149).

873 Other tests were conducted by Kern et al. (150) to find a nonlinear empirical model to

874 describe the adhesion that includes the load ( ), dwell time ( ), and separation rate ( ).

푙표푎푑 푑푤푒푙푙 푠푒푝 875 The main important parameters taken into consideration퐹 are the푇 maximum stress ( ), defined푣

876 as the ratio of the maximum measured force and the total capsule contact area achieved휎푚푎푥 during the

877 adhesion response, the total vertical probe displacement ( ) during the adhesion response, and

878 total effective adhesion energy ( ), defined as the total훿 area푡표푡푎푙 under the force displacement curve.

푒푓푓 879 The empirical equations are reported퐸 in Table 15. As the table shows, is a significant factor

880 only for and while has no observed effect. Moreover,퐹 푙표푎푑it has been noticed that as

푚푎푥 푒푓푓 푑푤푒푙푙 881 increases휎 퐸 and 푇 decrease.

푙표푎푑 푚푎푥 푒푓푓 882 퐹 휎 퐸 Table 15 : Adhesion model (150).

Critical design parameter Model equation

Maximum stress 0.31 2 푚푎푥 푠푒푝 푙표푎푑 푙표푎푑 푠푒푝 Total displacement 휎 = 972.491푣 − 7.711퐹 − 9.577퐹 푣 0.3 푡표푡푎푙 푠푒푝 Effective adhesion energy 훿 = 0.791푣 0.583 푒푓푓 푠푒푝 푙표푎푑 푠푒푝 883 Numerous studies have exploited mucoadhesion퐸 = (0.155 further푣 − by 0.010 developing퐹 푣 mucoadhesives) to

884 chemically bond to the mucosa to improve traction and/or adhesion. This can result in significantly

885 higher adhesive forces while, in some cases, maintaining the ability to repeatedly reattach to the

886 mucosa (151–154).

887 7.2 Active forces

888 The active forces on an object as it moves through the GI are shown with red arrows in

889 Figure 14 and encompass the tribology of the contact, the drag of the object moving through a

890 fluid, and tissue deformation during object motion. They depend largely on the properties of the

891 GI tract and the object, as well as the properties of the surrounding tissue and fluid. These forces

892 are applicable to all devices, including those with contact-based actuation, where the device must

893 maximize traction against the tissue; passive locomotion, where the device must minimize

894 frictional resistance to facilitate smooth passage through the lumen; and anchoring requirements,

895 where the device must secure itself, through high friction or adhesion, to the lumen.

896 7.2.1 Tribology 897 Figure 15 aims to summarize the primary factors affecting the tribology of an object

898 moving through the GI tract, including the size and shape of the object, its velocity, the properties

899 of the mucus, the properties of the tissue, the contact force, the texture of the object, and the texture

900 of the tissue. This is a complex interaction, with the mucosa and underlying tissue being

901 viscoelastic, inhomogeneous, and nonlinearly deformable (Chapter 5). Additionally, the contact

902 pressures – described in the previous section - vary considerably, as does the macro- and micro-

903 scale morphology of the tissue. The tissue also continually excretes a non–Newtonian mucus

904 (Section 3.5) which, depending on the velocity and scale of the device, can either increase or

905 decrease total frictional resistance.

906 907 908 Figure 15: Factors affecting the tribology of an object moving through the GI tract. 909 910 Understandably, it is complicated to create an all-encompassing and accurate model that

911 considers all the factors mentioned above. However, an understanding of the tribological

912 properties is useful to design functional surfaces and appropriate control techniques for this unique

913 environment. Sliker et al. developed a model to predict the resistance force on a capsule which

914 was validated by performing drag force experiments (155). Kim et al. developed an analytical

915 model based on a hoop stress analysis, and compared it to finite element model (FEM) results

916 (156). A similar model was developed by Woo et al. (157), using a hoop stress analysis and tensile

917 properties reported by Baek et al. (158), but including an empirical model for a propulsion force

918 due to electrical stimulus of the bowel.

919 Perhaps the most intuitive is a study by Zhang et al., where a velocity-dependent model is

920 presented (159). In this scenario, the total friction acting on a capsule can be written as a

921 summation of the environmental resistance, Coulomb friction, and viscous resistance (or drag)

922 (158,159)

923 ( )

푒 푣 푐 퐹 = 퐹 56+ 퐹 + 퐹 1

924 where is the environmental resistance, is the viscous resistance and is the Coulomb friction.

925 The environmental퐹푒 resistance is the amount퐹푣 of force required to deform퐹푐 the tissue in contact

926 with the object and is related to퐹 푒an elastic restoring force as

927 (2)

928 where θ퐹 푒is= the 푃 ·slant 푆 · 푠푖푛 angle (휃) of the object-tissue contact patch, P is contact pressure and S is the

929 contact area. Tissue is viscoelastic, and so increases with an increasing shear rate. This is shown

930 to be the dominant component of resistance푃 during an object’s interaction with the GI tissue and

931 has other names, including “edge effects” (160). The viscous friction or drag is related to the

932 rheological properties of the fluid in the contact patch and can be expressed as

933 (3)

푣 934 where the퐹 =apparent훿푣 viscosity coefficient, , d is the mean value of 휈 −0.7552 935 intestinal mucus thickness and is relative훿 velocity = 11.24 (159).(푑) In other+ 0. words,1148 this is the resistance of

936 the mucus during shear and is velocity푣 dependent. While static, resistance comes from the adhesive

937 bonds and, during shear, from the viscosity of the fluid (161,162). The Coulomb friction is decided

938 by

939 (4)

940 where 퐹푐 is= the µ · 푃coulomb · 푆 · 푐표푠 friction (휃) coefficient and the normal force has been replaced by

941 휇 to account for the hoop stress. The friction coefficient is influenced by the texture 푃of ·the 푆 ·

942 capsulecos (휃) and intestinal surface.

943 Equations (1) – (3) are all velocity dependent, and other literature supports this, while also

944 showing a total resistance dependency with object diameter, length (156,157,163,164), and normal

945 force (165,166). The key factors affecting the friction are the capsule dimensions, surface

946 geometry and the speed, while the effect of the weight is trivial. Ignoring the factor of weight, the 57

947 diameter affects the friction more than the length (167). Wang et al. (167) describe how resistance

948 changes with capsule size and velocity as

949 ( ) 1 ⁄푛 950 where 푓 (and푣) = 퐾푣 are+ related 퐶 to the R and L, radius and length of the capsule, respectively.5

951 퐾7.2.2 Fluid퐶 > 0drag 952 While there may not always be high volumes of fluid in the GI tract, it is important to

953 consider any impact of drag as an object moves through a fluid-filled environment. In these cases,

954 the drag opposes motion and is equal to

955 (6) 1 2 956 where 퐹 퐷is =the 2 density휌푣 퐶퐷 퐴of the fluid, is the velocity of the object, is the drag coefficient, and

957 is the contact휌 area of the front face 푣of the object. 퐶퐷 퐴

958

959 8. The impact of disease on GI physiology

960 Throughout this work we have described all the properties and characteristics of the GI

961 tract in its healthy state, which can be considered as the generic and most common condition.

962 However, having some knowledge of the possible GI alterations in the presence of digestive

963 diseases is useful, and in–depth investigation can be done as required for the application.

964 Therefore, here, we discuss the most common changes that can be seen from cancers and other

965 diseases, including IBS, IBD and celiac disease. IBD includes Ulcerative colitis (UC) and Crohn’s

966 disease (CD), both characterized by chronic inflammation of the gut (168). Although UC and CD

967 are grouped under IBD, they have different characteristics. UC is an inflammation condition of the

968 mucosa of the large intestine and is related to the presence of bacteria in the colon, which produce

969 colitis. However, CD usually occurs in the ileocaecal region (169). Both present an irregular

970 mucosal surface and transmucosal inflammation (169,170). Here we consider how these diseases

971 impact GI transit time, pH, microbiota and wall thickness (168).

972 Regarding the GI transit time, Bai et al. (168) report that there is no significant difference

973 in gastric emptying time and small intestine transit time between healthy subjects and IBS patients.

974 However, they report that the IBS patients have a longer colonic transit time. Regarding UC and

975 CD patients, Bai et al. report a slightly longer orocecal (mouth – cecum) transit time (168,171).

976 On–the–other–hand, celiac patients show a longer orocecal transit time but no alteration in small

977 intestine transit time (171).

978 Gastric and small intestine pH profiles in patients with IBD are similar to those in healthy

979 samples, while the pH of the CD colon is much lower (5.3 ± 0.3 in the right colon and 5.3 ± 0.7 in

980 the left colon) (172). Regarding celiac disease, a higher pH in the small bowel and unaltered pH

981 value in the stomach have been reported by Effinger et al. (171). Digestive diseases could also

982 mutate and reduce the intestinal concentration of bile salts, which affect the luminal pH and, thus,

983 the digestion of food (i.e. transit time) (168).

984 There is a strong correlation between gut microbiota, IBD, IBS and digestive diseases in

985 general. IBD has been shown to lead to a decrease of bacteria with anti–inflammatory capacities

986 (Proteobacteria) and an increase of bacteria with inflammatory capacities (Faecalibacterium,

987 Helicobacter species) (93,171). Regarding celiac patients, the microbiota was found to be rich in

988 potentially pathogenic bacteria and poor in species such as Lactobacilli and Bifidobacteria (171).

989 Regarding colorectal cancer, a study by Tojo et al. (173) shows that the alteration of composition

990 and function of the microbiota is correlated to the presence of colorectal cancer as well as IBD or

991 IBS. Sample of colorectal tumor have shown many bacterial such as Bacteroides vulgatus, E. coli

992 and Enterococcus faecalis (173). Other microbial systems have been reported in association with 59

993 gastroesophageal reflux (Veillonella, Prevotella, Haemophilus, Neisseria, Campylobacter, and

994 Fusobacterium) and adenocarcinoma in the esophagus (Campylobacter) (95). Moreover, gut

995 microbiota alterations may contribute to pancreatic diseases including pancreatitis, chronic

996 pancreatitis, and pancreatic cancer (92).

997 Wall thickness is a common indication of disease as it is proportional to the resulting

998 inflammation. With UC the small intestine is characteristically thickened and presents with

999 ulceration of the mucosa (169), while the colon wall can thicken up to 8 mm in the presence of CD

1000 (170). With regard to GI cancers, it has been proven that wall thickness is a good approach to

1001 evaluate and target the presence of a tumor. For example, in the esophagus a thickness above 5

1002 mm is considered abnormal (174). Similarly, Suk et al. (52) classify gastric diseases with respect

1003 to wall thickness. In particular, diseases have been classified as normal or benign disease (BD),

1004 early gastric cancer (EGC), and advanced gastric cancer (AGC). BD presents a thickness of the

1005 gastric wall of 4.9 ± 1.6 mm, the EGC shows a thickness of 5.6 ± 2.4 mm while a thickness of 10.3

1006 ± 4.7 mm is an indication of AGC.

1007

1008 9. Conclusion and future developments

1009 Considering that digestive disease can significantly impact the normal function of the GI

1010 tract and that GI cancer is one of the leading causes of death in the 21st century, the early diagnosis

1011 and subsequent treatment of GI disease is essential to reduce patient morbidity and mortality.

1012 Despite numerous technological advances in diagnosing and treating these diseases, the need for

1013 innovation still exists. This is partly due to the harsh, difficult–to–access environment that presents

1014 significant engineering challenges, but also to the increasing demand on health services by a

1015 growing population that has increasing disease prevalence. Therefore, there remains significant 60

1016 motivation for engineers in the biomedical field to find innovative, more sophisticated, and

1017 minimally invasive technologies to access the GI tract. In order to develop effective devices,

1018 engineers need a broad spectrum of knowledge on the GI system, and so in this review, the

1019 fundamental properties of the GI system – focusing on the esophagus, stomach, small and large

1020 intestines – were described with the goal of presenting key information.

1021 Developing disruptive medical devices for this region still has a number of major and open

1022 challenges. Firstly, the mechatronic design needs to be considered from the shape, dimensions,

1023 and materials of the device, to the research of innovative navigation strategies. The shape and

1024 dimensions must ensure safe and efficient passage through the tortuous and unstructured

1025 environment, while the material should be tailored to meet the friction, chemical resistance, and

1026 biocompatibility requirements (i.e. pH and microbiota of the environment). An innovative strategy

1027 for the device navigation is essential to ensure effective and real-time control and reduce the mean

1028 completion time of the procedure, which should at least be comparable with the existing procedure.

1029 This must be robust in an environment with numerous disturbances (i.e. respiration of the patient,

1030 peristaltic movements) and high variability between patients. To achieve this, localization,

1031 registration, and an effective locomotion mechanism (i.e. internal anchoring locomotion, external

1032 magnetic coupling locomotion or a novel hybrid combinations of internal and external locomotion)

1033 should be carefully considered depending on the context. Lastly, to bring added benefit, the device

1034 should provide effective diagnosis and or treatment. This should be accurately controlled with the

1035 device navigation and may be facilitated by context specific sensing, for example, combined time

1036 and pH measurements. In this context, the possibility of performing therapeutic functions, such as

1037 biopsy tissues, polyp ablation or drug delivery, is necessary. Therefore, enhanced and innovative

1038 devices have the potential to improve all these features, and thus, advance in the next decade, the

1039 medical and endoscopic field.

1040

1041

1042

1043

1044

1045

1046

1047

1048

1049

1050

1051

1052

1053

1054

1055

1056

1057

1058

1059

1060 List of Abbreviations:

1061 GI: gastrointestinal

1062 NIH: National Institutes of Health

1063 IBS: irritable bowel syndrome

1064 IBD: inflammable bowel disease

1065 CD: Crohn’s disease

1066 UC: ulcerative colitis

1067 IR: incidence rate

1068 MR: mortality rate

1069 CT: computed tomography

1070 MRI: magnetic resonance imaging

1071 CE: endoscopic capsule

1072 VAS: visual analogue scale

1073 ICC: interstitial cells of Cajal

1074 RPC: rhythmic phasic contraction

1075 GMC: giant migrating contraction

1076 MMC: migratory motor complex

1077 IAP: intra-abdominal pressure

1078 MFS: migrating motor complex force sensor

1079 FEM: finite element model

1080

1081

1082 References

1083 1. Caffrey CM, Chevalerias O, O’Mathuna C, Twomey K. Swallowable-Capsule Technology. 1084 IEEE Pervasive Computing. 2008 Jan;7(1):23–9.

1085 2. Health Information [Internet]. National Institutes of Health (NIH). [cited 2019 Mar 13]. 1086 Available from: https://www.nih.gov/health-information

1087 3. Tsai S, Wendt J, Cardarelli K, Fraser A. A mortality and morbidity study of refinery and 1088 petrochemical employees in Louisiana. Occup Environ Med. 2003 Sep;60(9):627–33.

1089 4. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 1090 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 1091 countries. CA: A Cancer Journal for Clinicians. 2018 Nov;68(6):394–424.

1092 5. Valdastri P, Ciuti G, Verbeni A, Menciassi A, Dario P, Arezzo A, et al. Magnetic air capsule 1093 robotic system: Proof of concept of a novel approach for painless colonoscopy. Surgical 1094 Endoscopy and Other Interventional Techniques. 2012;26(5):1238–1246.

1095 6. CDC. CDC VitalSigns - Colorectal Cancer [Internet]. Centers for Disease Control and 1096 Prevention. 2011 [cited 2019 Mar 13]. Available from: 1097 https://www.cdc.gov/vitalsigns/cancerscreening/colorectalcancer/index.html

1098 7. Joseph DA, Degroff AS, Hayes NS, Wong FL, Plescia M. The Colorectal Cancer Control 1099 Program: Partnering to increase population level screening. Gastrointestinal Endoscopy. 1100 2011;73(3):429–434.

1101 8. Bujanda L, Sarasqueta C, Zubiaurre L, Cosme A, Muñoz C, Sánchez A, et al. Low 1102 adherence to colonoscopy in the screening of first-degree relatives of patients with 1103 colorectal cancer. Gut. 2007;56(12):1714–1718.

1104 9. Obstein KL. Advanced endoscopic technologies for colorectal cancer screening. World 1105 Journal of Gastroenterology. 2013;19(4):431.

1106 10. Valdastri P, Simi M, Webster RJ. Advanced Technologies for Gastrointestinal Endoscopy. 1107 Annual Review of Biomedical Engineering. 2012;14(1):397–429.

1108 11. Li B, Meng MQH, Lau JYW. Computer-aided small bowel tumor detection for capsule 1109 endoscopy. Artificial Intelligence in Medicine. 2011;52(1):11–16.

1110 12. Brambs H-J, Juchems MS. Virtual endoscopy using CT scan. Minimally Invasive Therapy 1111 & Allied Technologies. 2003 Jan;12(5):207–16.

1112 13. Kumar RR, Joseph J, Vidya PV, Pournami C, John NJ. Virtual colonoscopy: A plausible 1113 alternative to conventional colonoscopy. In: 2017 IEEE Region 10 Symposium 1114 (TENSYMP) [Internet]. Cochin, India: IEEE; 2017 [cited 2020 Jun 12]. p. 1–5. Available 1115 from: http://ieeexplore.ieee.org/document/8070107/ 64

1116 14. Slawinski PR, Obstein KL, Valdastri P. Emerging issues and future developments in capsule 1117 endoscopy. Techniques in Gastrointestinal Endoscopy. 2015 Jan 1;17(1):40–6.

1118 15. Liao Z, Hou X, Lin-Hu E-Q, Sheng J-Q, Ge Z-Z, Jiang B, et al. Accuracy of Magnetically 1119 Controlled Capsule Endoscopy, Compared With Conventional Gastroscopy, in Detection of 1120 Gastric Diseases. Clinical Gastroenterology and Hepatology. 2016 Sep;14(9):1266- 1121 1273.e1.

1122 16. Nam S-J, Lee HS, Lim YJ. Evaluation of Gastric Disease with Capsule Endoscopy. Clinical 1123 Endoscopy. 2018 Jul 31;51(4):323–8.

1124 17. Riccioni ME. Colon capsule endoscopy: Advantages, limitations and expectations. Which 1125 novelties? World Journal of Gastrointestinal Endoscopy. 2012;4(4):99.

1126 18. Slawinski PR. Capsule endoscopy of the future: What’s on the horizon? World Journal of 1127 Gastroenterology. 2015;21(37):10528.

1128 19. Ciuti G, Menciassi A, Dario P. Capsule Endoscopy: From Current Achievements to Open 1129 Challenges. IEEE Rev Biomed Eng. 2011;4:59–72.

1130 20. Liu L, Towfighian S, Hila A. A Review of Locomotion Systems for Capsule Endoscopy. 1131 IEEE Reviews in Biomedical Engineering. 2015;8:138–51.

1132 21. Pittiglio G, Barducci L, Martin JW, Norton J, Avizzano CA, Obstein K, et al. Magnetic 1133 Levitation for Soft-Tethered Capsule Colonoscopy Actuated with a Single Permanent 1134 Magnet: a Dynamic Control Approach. IEEE Robotics and Automation Letters. 2019;1–1.

1135 22. Zhang P, Li J, Hao Y, Ciuti G, Arai T, Huang Q, et al. A compensation strategy for accurate 1136 orientation of a tethered robotic capsule endoscope. In: 2017 IEEE International Conference 1137 on Cyborg and Bionic Systems (CBS) [Internet]. Beijing: IEEE; 2017 [cited 2020 Jun 12]. 1138 p. 257–61. Available from: http://ieeexplore.ieee.org/document/8266111/

1139 23. Paul Glass, Metin Sitti, Arjun Pennathur, Ragunath Appasamy. A Swallowable Tethered 1140 Capsule Endoscope for Diagnosing Barrett’s Esophagus. GIE [Internet]. 2009 Apr 1;69(5). 1141 Available from: https://www.giejournal.org/article/S0016-5107(09)00561-6/fulltext

1142 24. Gora MJ, Sauk JS, Carruth RW, Gallagher KA, Suter MJ, Nishioka NS, et al. Tethered 1143 capsule endomicroscopy enables less invasive imaging of gastrointestinal tract 1144 microstructure. Nature Medicine. 2013;19(2):238–240.

1145 25. Caprara R, Obstein KL, Scozzarro G, Natali CD, Beccani M, Morgan DR, et al. A Platform 1146 for Gastric Cancer Screening in Low- and Middle-Income Countries. IEEE Trans Biomed 1147 Eng. 2015 May;62(5):1324–32.

1148 26. Dehghani H, Welch CR, Pourghodrat A, Nelson CA, Oleynikov D, Dasgupta P, et al. Design 1149 and preliminary evaluation of a self-steering, pneumatically driven colonoscopy robot. 1150 Journal of Medical Engineering and Technology. 2017;41(3):223–236. 65

1151 27. Mapara SS, Patravale VB. Medical capsule robots: A renaissance for diagnostics, drug 1152 delivery and surgical treatment. Journal of Controlled Release. 2017 Sep;261:337–51.

1153 28. Toennies JL, Tortora G, Simi M, Valdastri P, Webster RJ. Swallowable medical devices for 1154 diagnosis and surgery: The state of the art. Proceedings of the Institution of Mechanical 1155 Engineers, Part C: Journal of Mechanical Engineering Science. 2010 Jul;224(7):1397–414.

1156 29. Valdastri P, Quaglia C, Susilo E, Menciassi A, Dario P, Ho C, et al. Wireless therapeutic 1157 endoscopic capsule: in vivo experiment. Endoscopy. 2008 Dicembre;40(12):979–982.

1158 30. Beccani M, Di Natali C, Aiello G, Benjamin C, Susilo E, Valdastri P. A Magnetic Drug 1159 Delivery Capsule Based on a Coil Actuation Mechanism. Procedia Engineering. 1160 2015;120:53–6.

1161 31. Kong KC, Cha J, Jeon D, Cho DID. A rotational micro biopsy device for the capsule 1162 endoscope. 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 1163 IROS. 2005;3057–3061.

1164 32. Norton JC, Slawinski PR, Lay HS, Martin JW, Cox BF, Cummins G, et al. Intelligent 1165 magnetic manipulation for gastrointestinal ultrasound. Sci Robot. 2019 Jun 1166 19;4(31):eaav7725.

1167 33. Hounnou G, Destrieux C, Desmé J, Bertrand P, Velut S. Anatomical study of the length of 1168 the human intestine. Surgical and Radiologic Anatomy. 2002;24(5):290–294.

1169 34. McKay DM, Bienenstock J. The interaction between mast cells and nerves in the 1170 gastrointestinal tract. Immunology Today. 1994 Nov 1;15(11):533–8.

1171 35. Soybel DI. Anatomy and Physiology of the Stomach. Surgical Clinics. 2005 Oct 1172 1;85(5):875–94.

1173 36. Gelberg HB. Comparative Anatomy, Physiology, and Mechanisms of Disease Production 1174 of the Esophagus, Stomach, and Small Intestine. Toxicologic Pathology. 2014 1175 Jan;42(1):54–66.

1176 37. Helander HF, Fändriks L. Surface area of the digestive tract – revisited. Scandinavian 1177 Journal of Gastroenterology. 2014 Jun;49(6):681–9.

1178 38. Ødegaard S, Nesje LB, Lærum OD, Kimmey MB. High-frequency ultrasonographic 1179 imaging of the gastrointestinal wall. Expert Review of Medical Devices. 2012 1180 May;9(3):263–73.

1181 39. Daniela Urma BK. Esophagus - anatomy and development. GI Motility online (2006) 1182 [Internet]. 2006 May 16; Available from: 1183 https://www.nature.com/gimo/contents/pt1/full/gimo6.html

1184 40. Wolfram T Knoefel PA. Swallowed Foreign Bodies in Adults. Available from: 1185 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3536040/

1186 41. Ikenberry SO, Jue TL, Anderson MA, Appalaneni V, Banerjee S, Ben-Menachem T, et al. 1187 Management of ingested foreign bodies and food impactions. Gastrointestinal Endoscopy. 1188 2011 Jun;73(6):1085–91.

1189 42. Bekkerman M, Sachdev AH, Andrade J, Twersky Y, Iqbal S. Endoscopic Management of 1190 Foreign Bodies in the Gastrointestinal Tract: A Review of the Literature. Gastroenterology 1191 Research and Practice. 2016;2016:1–6.

1192 43. Xia F, Mao J, Ding J, Yang H. Observation of normal appearance and wall thickness of 1193 esophagus on CT images. European Journal of Radiology. 2009 Dec;72(3):406–11.

1194 44. Moss A, Schnyder P, Thoeni R, Margulis A. Esophageal carcinoma: pretherapy staging by 1195 computed tomography. American Journal of Roentgenology. 1981 Jun;136(6):1051–6.

1196 45. Diana C. Peterson SRC. Anatomy, Abdomen and Pelvis, Stomach. Available from: 1197 https://www.ncbi.nlm.nih.gov/books/NBK482334/

1198 46. ramkumar d., schulze k. s. The pylorus. Neurogastroenterology and Motility. 2005 1199 Jun;17(s1):22–30.

1200 47. Doran S, Jones KL, Andrews JM, Horowitz M. Effects of meal volume and posture on 1201 gastric emptying of solids and appetite. American Journal of Physiology-Regulatory, 1202 Integrative and Comparative Physiology. 1998;275(5):R1712–R1718.

1203 48. Moore JG, Datz FL, Christian PE, Greenberg E, Alazraki N. Effect of body posture on 1204 radionuclide measurements of gastric emptying. Digestive Diseases and Sciences. 1205 1988;33(12):1592–1595.

1206 49. Sanaka M, Anjiki H, Tsutsumi H, Abe K, Kawakami T, Saitoh M, et al. Effect of cigarette 1207 smoking on gastric emptying of solids in Japanese smokers: a crossover study using the 1208 13C-octanoic acid breath test. Journal of Gastroenterology. 2005 Jul 8;40(6):578–82.

1209 50. Bennink R, Peeters M, Van den Maegdenbergh V, Geypens B, Rutgeerts P, De Roo M, et 1210 al. Comparison of total and compartmental gastric emptying and antral motility between 1211 healthy men and women. European journal of nuclear medicine. 1998;25(9):1293–1299.

1212 51. Vasavid P, Chaiwatanarat T, Pusuwan P, Sritara C, Roysri K, Namwongprom S, et al. 1213 Normal Solid Gastric Emptying Values Measured by Scintigraphy Using Asian-style 1214 Meal:A Multicenter Study in Healthy Volunteers. J Neurogastroenterol Motil. 2014 1215 Jul;20(3):371–8.

1216 52. Suk KT, Lim DW, Kim MY, Park DH, Kim KH, Kim JM, et al. Thickening of the gastric 1217 wall on transabdominal sonography: A sign of gastric cancer. Journal of Clinical 1218 Ultrasound. 2008 Oct;36(8):462–6. 67

1219 53. Huh CH, Bhutani MS, Farfán EB, Bolch WE. Individual variations in mucosa and total wall 1220 thickness in the stomach and rectum assessed via endoscopic ultrasound. Physiol Meas. 1221 2003 Nov;24(4):N15-22.

1222 54. Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of 1223 humans and commonly used laboratory animals. Biopharmaceutics & Drug Disposition. 1224 1995 Jul 1;16(5):351–80.

1225 55. Madhu Badireddy JTC. Anatomy, Abdomen and Pelvis, Small Intestine. Available from: 1226 https://www.ncbi.nlm.nih.gov/books/NBK459366/

1227 56. Meyers MA. Treitz redux: the ligament of Treitz revisited. Abdom Imaging. 1995 Sep 1228 1;20(5):421–4.

1229 57. Ellis H. Anatomy of the small intestine (jejunum and ileum). Surgery (Oxford). 2011 1230 Aug;29(8):355–7.

1231 58. Weaver LT, Austin S, Cole TJ. Small intestinal length: a factor essential for gut adaptation. 1232 Gut. 1991 Nov 1;32(11):1321–3.

1233 59. Hosseinpour M, Behdad A. Evaluation of small bowel measurement in alive patients. 1234 Surgical and Radiologic Anatomy. 2008 Nov;30(8):653–5.

1235 60. Uro DD. Overview of short bowel syndrome and intestinal transplantation. Colombia 1236 Médica. 2007;38:5.

1237 61. Cronin CG, Delappe E, Lohan DG, Roche C, Murphy JM. Normal small bowel wall 1238 characteristics on MR enterography. European Journal of Radiology. 2010 Aug 1239 1;75(2):207–11.

1240 62. stany lobo PK. Anatomy, Abdomen and Pelvis, Large Intestine. Available from: 1241 https://www.ncbi.nlm.nih.gov/books/NBK470577/

1242 63. Eickhoff A, Pickhardt PJ, Hartmann D, Riemann JF. Colon anatomy based on CT 1243 colonography and fluoroscopy: Impact on looping, straightening and ancillary manoeuvres 1244 in colonoscopy. Digestive and Liver Disease. 2010 Apr;42(4):291–6.

1245 64. Alazmani A, Hood A, Jayne D, Neville A, Culmer P. Quantitative assessment of colorectal 1246 morphology: Implications for robotic colonoscopy. Medical Engineering and Physics. 1247 2016;38(2):148–154.

1248 65. Sadahiro S, Ohmura T, Yamada Y, Saito T, Taki Y. Analysis of length and surface area of 1249 each segment of the large intestine according to age, sex and physique. Surgical and 1250 Radiologic Anatomy. 1992 Sep;14(3):251–7.

1251 66. Wiesner W, Mortelé KJ, Ji H, Ros PR. Normal Colonic Wall Thickness at CT and Its 1252 Relation to Colonic Distension: Journal of Computer Assisted Tomography. 2002 1253 Jan;26(1):102–6.

1254 67. Trefts E, Gannon M, Wasserman DH. The liver. Current Biology. 2017 Nov;27(21):R1147– 1255 51.

1256 68. Turumin JL, Shanturov VA, Turumina HE. The role of the gallbladder in humans. Revista 1257 de Gastroenterología de México. 2013 Jul;78(3):177–87.

1258 69. Anatomy and Histology of the Pancreas [Internet]. The University of Michigan Library; 1259 [cited 2020 Jan 7]. Available from: http://pancreapedia.org/?q=node/8098

1260 70. Coffey JC. Surgical anatomy and anatomic surgery–clinical and scientific mutualism. The 1261 Surgeon. 2013;11(4):177–182.

1262 71. Byrnes KG, Walsh D, Dockery P, McDermott K, Coffey JC. Anatomy of the mesentery: 1263 Current understanding and mechanisms of attachment. Seminars in Cell & Developmental 1264 Biology [Internet]. 2018 Oct 15 [cited 2018 Dec 17]; Available from: 1265 http://www.sciencedirect.com/science/article/pii/S1084952118302052

1266 72. White EJ, Cunnane EM, McMahon M, Walsh MT, Coffey JC, O’Sullivan L. Mechanical 1267 characterisation of porcine non-intestinal colorectal tissues for innovation in surgical 1268 instrument design. Proceedings of the Institution of Mechanical Engineers, Part H: Journal 1269 of Engineering in Medicine. 2018 Aug;232(8):796–806.

1270 73. Canadian Cancer Society [Internet]. Available from: www.cancer.ca

1271 74. Kimmey MB, Martin RW, Haggitt RC, Wang KY, Franklin DW, Silverstein FE. Histologic 1272 Correlates of Gastrointestinal Ultrasound Images. Gastroenterology. 1989 Feb;96(2):433– 1273 41.

1274 75. Nylund K, Hausken T, Ødegaard S, Eide G, Gilja O. Gastrointestinal Wall Thickness 1275 Measured with Transabdominal Ultrasonography and Its Relationship to Demographic 1276 Factors in Healthy Subjects. Ultraschall in der Medizin - European Journal of Ultrasound. 1277 2012 Apr 13;33(07):E225–32.

1278 76. Marchesini R, Pignoli E, Tomatis S, Fumagalli S, Sichirollo AE, Palma SD, et al. Ex vivo 1279 optical properties of human colon tissue. Lasers in Surgery and Medicine. 1994;15(4):351– 1280 7.

1281 77. Lai SK, Wang Y-Y, Wirtz D, Hanes J. Micro- and macrorheology of mucus. Advanced 1282 Drug Delivery Reviews. 2009 Feb;61(2):86–100.

1283 78. Pelaseyed T, Bergström JH, Gustafsson JK, Ermund A, Birchenough GMH, Schütte A, et 1284 al. The mucus and mucins of the goblet cells and enterocytes provide the first defense line

1285 of the gastrointestinal tract and interact with the immune system. Immunological Reviews. 1286 2014 Jul;260(1):8–20.

1287 79. Cone RA. Barrier properties of mucus. Advanced Drug Delivery Reviews. 2009 1288 Feb;61(2):75–85.

1289 80. Leal J, Smyth HDC, Ghosh D. Physicochemical properties of mucus and their impact on 1290 transmucosal drug delivery. International Journal of Pharmaceutics. 2017 Oct;532(1):555– 1291 72.

1292 81. Lock JY, Carlson TL, Carrier RL. Mucus models to evaluate the diffusion of drugs and 1293 particles. Advanced Drug Delivery Reviews. 2018 Jan;124:34–49.

1294 82. Boegh M, Nielsen HM. Mucus as a Barrier to Drug Delivery - Understanding and 1295 Mimicking the Barrier Properties. Basic & Clinical Pharmacology & Toxicology. 2015 1296 Mar;116(3):179–86.

1297 83. Johansson MEV, Sjövall H, Hansson GC. The gastrointestinal mucus system in health and 1298 disease. Nature Reviews Gastroenterology & Hepatology. 2013 Jun;10(6):352–61.

1299 84. Gustafsson JK, Ermund A, Johansson MEV, Schütte A, Hansson GC, Sjövall H. An ex vivo 1300 method for studying mucus formation, properties, and thickness in human colonic biopsies 1301 and mouse small and large intestinal explants. American Journal of Physiology- 1302 Gastrointestinal and Liver Physiology. 2012 Feb;302(4):G430–8.

1303 85. Lindahl A, Ungell A-L, Knutson L, Lennernäs H. Characterization of Fluids from the 1304 Stomach and Proximal Jejunum in Men and Women. Pharm Res. 1997 Apr 1;14(4):497– 1305 502.

1306 86. Evans DF, Pye G, Bramley R, Clark AG, Dyson TJ, Hardcastle JD. Measurement of 1307 gastrointestinal pH profiles in normal ambulant human subjects. Gut. 1988 Aug 1308 1;29(8):1035–41.

1309 87. McDougall CJ, Wong R, Scudera P, Lesser M, DeCosse JJ. Colonic mucosal pH in humans. 1310 Digest Dis Sci. 1993 Mar 1;38(3):542–5.

1311 88. Donald O. Castell RT. Gastroesophageal reflux monitoring: pH and impedance. GI Motility 1312 online (2006) [Internet]. 2006 May 16; Available from: 1313 https://www.nature.com/gimo/contents/pt1/full/gimo31.html

1314 89. Singh H, Ye A, Horne D. Structuring food emulsions in the gastrointestinal tract to modify 1315 lipid digestion. Progress in Lipid Research. 2009 Mar;48(2):92–100.

1316 90. Boland M. Human digestion - a processing perspective. Journal of the Science of Food and 1317 Agriculture. 2016 May;96(7):2275–83.

1318 91. C. E. Stevenson RLW. Chemical Digestion and Absorption: A Closer Look. In. Available 1319 from: https://opentextbc.ca/anatomyandphysiology/chapter/23-7-chemical-digestion-and- 1320 absorption-a-closer-look/

1321 92. Akshintala VS, Talukdar R, Singh VK, Goggins M. The Gut Microbiome in Pancreatic 1322 Disease. Clinical Gastroenterology and Hepatology. 2019 Jan;17(2):290–5.

1323 93. Nishida A, Inoue R, Inatomi O, Bamba S, Naito Y, Andoh A. Gut microbiota in the 1324 pathogenesis of inflammatory bowel disease. Clin J Gastroenterol. 2018 Feb;11(1):1–10.

1325 94. Ley RE, Peterson DA, Gordon JI. Ecological and evolutionary forces shaping microbial 1326 diversity in the human intestine. Cell. 2006;124(4):837–848.

1327 95. Di Pilato V, Freschi G, Ringressi MN, Pallecchi L, Rossolini GM, Bechi P. The esophageal 1328 microbiota in health and disease: Esophageal microbiota and disease. Annals of the New 1329 York Academy of Sciences. 2016 Oct;1381(1):21–33.

1330 96. Bik EM, Eckburg PB, Gill SR, Nelson KE, Purdom EA, Francois F, et al. Molecular analysis 1331 of the bacterial microbiota in the human stomach. Proceedings of the National Academy of 1332 Sciences. 2006 Jan 17;103(3):732–7.

1333 97. Zoetendal EG, Raes J, van den Bogert B, Arumugam M, Booijink CC, Troost FJ, et al. The 1334 human small intestinal microbiota is driven by rapid uptake and conversion of simple 1335 carbohydrates. ISME J. 2012 Jul;6(7):1415–26.

1336 98. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, et al. 1337 Microbiology: Diversity of the human intestinal microbial flora. Science. 1338 2005;308(5728):1635–1638.

1339 99. Jandhyala SM. Role of the normal gut microbiota. World Journal of Gastroenterology. 1340 2015;21(29):8787.

1341 100. Krishnan S, Alden N, Lee K. Pathways and functions of gut microbiota metabolism 1342 impacting host physiology. Current Opinion in Biotechnology. 2015 Dec;36:137–45.

1343 101. Egorov VI, Schastlivtsev IV, Prut EV, Baranov AO, Turusov RA. Mechanical properties of 1344 the human gastrointestinal tract. Journal of Biomechanics. 2002 Oct 1;35(10):1417–25.

1345 102. Lim Y-J, Deo D, Singh TP, Jones DB, De S. In situ measurement and modeling of 1346 biomechanical response of human cadaveric soft tissues for physics-based surgical 1347 simulation. Surgical endoscopy. 2009;23(6):1298.

1348 103. Egorov VI, Schastlivtsev V, Turusov RA, Baranov AO. Participation of the intestinal layers 1349 in supplying of the mechanical strength of the intact and sutured gut. Eur Surg Res. 2002 1350 Dec;34(6):425–31.

1351 104. Vanags I, Petersons A, Ose V, Ozolanta I, Kasyanov V, Laizans J, et al. Biomechanical 1352 properties of oesophagus wall under loading. Journal of Biomechanics. 2003 Sep 1353 1;36(9):1387–90.

1354 105. Yang W, Fung TC, Chian KS, Chong CK. Viscoelasticity of Esophageal Tissue and 1355 Application of a QLV Model. J Biomech Eng. 2006 May 11;128(6):909–16.

1356 106. Jia ZG, Li W, Zhou ZR. Mechanical characterization of stomach tissue under uniaxial 1357 tensile action. Journal of Biomechanics. 2015 Feb 26;48(4):651–8.

1358 107. Tan R, Liu H, Su G, Zhang C, Li H, Wang Y. Experimental investigation of the small 1359 intestine’s viscoelasticity for the motion of capsule robot. In: 2011 IEEE International 1360 Conference on Mechatronics and Automation [Internet]. Beijing, China: IEEE; 2011 [cited 1361 2020 Mar 4]. p. 249–53. Available from: http://ieeexplore.ieee.org/document/5985665/

1362 108. O’Connor A, O’Moráin C. Digestive Function of the Stomach. DDI. 2014;32(3):186–91.

1363 109. Sanders KM, Koh SD, Ro S, Ward SM. Regulation of gastrointestinal motility-insights from 1364 smooth muscle biology. Nature Reviews Gastroenterology and Hepatology. 1365 2012;9(11):633–645.

1366 110. Takaki M. Gut Pacemaker Cells: the Interstitial Cells of Cajal (ICC). Journal of Smooth 1367 Muscle Research. 2003;39(5):137–161.

1368 111. Hinder RA, Kelly KA. Human gastric pacesetter potential. The American Journal of 1369 Surgery. 1977 Jan;133(1):29–33.

1370 112. O’Grady G, Du P, Egbuji JU, Lammers WJEP, Wahab A, Pullan AJ, et al. A novel 1371 laparoscopic device for the measurement of gastrointestinal slow wave activity. Surg 1372 Endosc [Internet]. 2009 Dec [cited 2018 Aug 25];23(12):2842–8. Available from: 1373 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4106912/

1374 113. Cheng LK. Slow wave conduction patterns in the stomach: from Waller’s Foundations to 1375 Current Challenges. Acta Physiologica. 2015 Feb;213(2):384–93.

1376 114. Vilarino F, Spyridonos P, DeIorio F, Vitria J, Azpiroz F, Radeva P. Intestinal Motility 1377 Assessment With Video Capsule Endoscopy: Automatic Annotation of Phasic Intestinal 1378 Contractions. IEEE Transactions on Medical Imaging. 2010 Feb;29(2):246–59.

1379 115. Vu H, Echigo T, Sagawa R, Yagi K, Shiba M, Higuchi K, et al. Detection of contractions 1380 in adaptive transit time of the small bowel from wireless capsule endoscopy videos. 1381 Computers in Biology and Medicine [Internet]. 2009 Jan [cited 2018 Aug 25];39(1):16–26. 1382 Available from: http://linkinghub.elsevier.com/retrieve/pii/S0010482508001522

1383 116. Fleckenstein P. Migrating electrical spike activity in the fasting human small intestine. The 1384 American Journal of Digestive Diseases. 1978 Sep;23(9):769–75.

1385 117. Sarna SK. Colonic Motility in Health [Internet]. Morgan & Claypool Life Sciences; 2010 1386 [cited 2019 Mar 12]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK53471/

1387 118. Rao SSC, Sadeghi P, Beaty J, Kavlock R, Ackerson K. Ambulatory 24-h colonic manometry 1388 in healthy humans. American Journal of Physiology-Gastrointestinal and Liver Physiology. 1389 2001 Apr;280(4):G629–39.

1390 119. O’Grady G, Du P, Cheng LK, Egbuji JU, Lammers WJEP, Windsor JA, et al. Origin and 1391 propagation of human gastric slow-wave activity defined by high-resolution mapping. 1392 American Journal of Physiology-Gastrointestinal and Liver Physiology. 2010 1393 Sep;299(3):G585–92.

1394 120. Bassotti G, Gaburri M. Manometric investigation of high-amplitude propagated contractile 1395 activity of the human colon. American Journal of Physiology-Gastrointestinal and Liver 1396 Physiology. 1988 Nov 1;255(5):G660–4.

1397 121. Worsøe J, Fynne L, Gregersen T, Schlageter V, Christensen LA, Dahlerup JF, et al. Gastric 1398 transit and small intestinal transit time and motility assessed by a magnet tracking system. 1399 BMC Gastroenterology [Internet]. 2011 Dec [cited 2018 Aug 22];11(1). Available from: 1400 http://bmcgastroenterol.biomedcentral.com/articles/10.1186/1471-230X-11-145

1401 122. Degen LP, Phillips SF. Variability of gastrointestinal transit in healthy women and men. 1402 Gut. 1996 Aug 1;39(2):299–305.

1403 123. Maurer AH. Gastrointestinal Motility, Part 2: Small-Bowel and Colon Transit. J Nucl Med 1404 [Internet]. 2015 Sep 1 [cited 2018 Sep 30];56(9):1395–400. Available from: 1405 http://jnm.snmjournals.org/content/56/9/1395

1406 124. Miller MA, Parkman HP, Urbain JL, Brown KL, Donahue DJ, Knight LC, et al. Comparison 1407 of scintigraphy and lactulose breath hydrogen test for assessment of orocecal transit: 1408 lactulose accelerates small bowel transit. Dig Dis Sci. 1997 Jan;42(1):10–8.

1409 125. Camilleri M, Colemont LJ, Phillips SF, Brown ML, Thomforde GM, Chapman N, et al. 1410 Human gastric emptying and colonic filling of solids characterized by a new method. 1411 American Journal of Physiology-Gastrointestinal and Liver Physiology. 1412 1989;257(2):G284–G290.

1413 126. Krevsky B, Maurer AH, Niewiarowski T, Cohen S. Effect of verapamil on human intestinal 1414 transit. Digestive Diseases and Sciences. 1992 Jun;37(6):919–24.

1415 127. Cummings JH, Jenkins DJ, Wiggins HS. Measurement of the mean transit time of dietary 1416 residue through the human gut. Gut [Internet]. 1976 Mar [cited 2019 Mar 24];17(3):210–8. 1417 Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1411154/

1418 128. Deloose E, Janssen P, Depoortere I, Tack J. The migrating motor complex: control 1419 mechanisms and its role in health and disease. Nature Reviews Gastroenterology &

1420 Hepatology [Internet]. 2012 May [cited 2018 Aug 27];9(5):271–85. Available from: 1421 https://www.nature.com/articles/nrgastro.2012.57

1422 129. Takahashi T. Interdigestive migrating motor complex -its mechanism and clinical 1423 importance. Journal of Smooth Muscle Research. 2013;49(0):99–111.

1424 130. Jouet P, Coffin B, Lémann M, Gorbatchef C, Franchisseur C, Jian R, et al. Tonic and phasic 1425 motor activity in the proximal and distal colon of healthy humans. American Journal of 1426 Physiology-Gastrointestinal and Liver Physiology. 1998 Mar;274(3):G459–64.

1427 131. Ferrua M, Singh R. Modeling the Fluid Dynamics in a Human Stomach to Gain Insight of 1428 Food Digestion. J Food Sci. 2010 Sep;75(7):R151–62.

1429 132. Milo R, Jorgensen P, Moran U, Weber G, Springer M. BioNumbers—the database of key 1430 numbers in molecular and cell biology. Nucleic Acids Research. 2010 1431 Jan;38(suppl_1):D750–3.

1432 133. Malbrain MLNG, De laet IE, De Waele JJ, Kirkpatrick AW. Intra-abdominal hypertension: 1433 Definitions, monitoring, interpretation and management. Best Practice & Research Clinical 1434 Anaesthesiology. 2013 Jun;27(2):249–70.

1435 134. Balogh Z, Jones F, D’Amours S, Parr M, Sugrue M. Continuous intra-abdominal pressure 1436 measurement technique. The American Journal of Surgery. 2004 Dec;188(6):679–84.

1437 135. De Keulenaer BL, De Waele JJ, Powell B, Malbrain MLNG. What is normal intra- 1438 abdominal pressure and how is it affected by positioning, body mass and positive end- 1439 expiratory pressure? Intensive Care Medicine. 2009 Jun;35(6):969–76.

1440 136. Neumann P, Gill V. Pelvic Floor and Abdominal Muscle Interaction: EMG Activity and 1441 Intra-abdominal Pressure. International Urogynecology Journal. 2002 Apr;13(2):125–32.

1442 137. Miftahof R, Akhmadeev N. Dynamics of intestinal propulsion. Journal of theoretical 1443 biology. 2007;246(2):377–393.

1444 138. Arshak A, Arshak K, Waldron D, Morris D, Korostynska O, Jafer E, et al. Review of the 1445 potential of a wireless MEMS and TFT microsystems for the measurement of pressure in 1446 the GI tract. Medical engineering & physics. 2005;27(5):347–356.

1447 139. Bertuzzi A, Salinari S, Mancinelli R, Pescatori M. Peristaltic transport of a solid bolus. 1448 Journal of biomechanics. 1983;16(7):459–464.

1449 140. Miftahof R, Fedotov E. Intestinal propulsion of a solid non-deformable bolus. Journal of 1450 Theoretical Biology. 2005;235(1):57–70.

1451 141. Miftahof RN. The wave phenomena in smooth muscle syncytia. In silico biology. 2005;5(5, 1452 6):479–498.

1453 142. Terry BS, Schoen JA, Rentschler ME. Characterization and experimental results of a novel 1454 sensor for measuring the contact force from myenteric contractions. IEEE Transactions on 1455 Biomedical Engineering. 2012;59(7):1971–1977.

1456 143. Terry BS, Schoen JA, Rentschler ME. Measurements of the contact force from myenteric 1457 contractions on a solid bolus. Journal of robotic surgery. 2013;7(1):53–57.

1458 144. Terry BS, Francisco MM, Schoen JA, Rentschler ME. Sensor for Measuring the Contact 1459 Force From Human Myenteric Contractions for In Vivo Robotic Capsule Endoscope 1460 Mobility. J Med Devices. 2013 Sep 1;7(3):030911.

1461 145. Terry BS, Lyle AB, Schoen JA, Rentschler ME. Preliminary Mechanical Characterization 1462 of the Small Bowel for In Vivo Robotic Mobility. Journal of Biomechanical Engineering. 1463 2011;133(9):091010.

1464 146. Calio R, Camboni D, Alcaide JO, Oddo CM, Carrozza MC, Menciassi A, et al. Robotic 1465 endoscopic capsule for closed-loop force-based control and safety strategies. In: 2017 IEEE 1466 International Conference on Cyborg and Bionic Systems (CBS) [Internet]. Beijing: IEEE; 1467 2017 [cited 2018 Oct 4]. p. 253–6. Available from: 1468 http://ieeexplore.ieee.org/document/8266110/

1469 147. King M. Experimental models for studying mucociliary clearance. European Respiratory 1470 Journal. 1998;11(1):222–228.

1471 148. Terry BS, Passernig AC, Hill ML, Schoen JA, Rentschler ME. Small intestine mucosal 1472 adhesivity to in vivo capsule robot materials. Journal of the mechanical behavior of 1473 biomedical materials. 2012;15:24–32.

1474 149. Varum FJO, Veiga F, Sousa JS, Basit AW. An investigation into the role of mucus thickness 1475 on mucoadhesion in the gastrointestinal tract of pig. European Journal of Pharmaceutical 1476 Sciences. 2010 Jul;40(4):335–41.

1477 150. Kern MD, Ortega Alcaide J, Rentschler ME. Soft material adhesion characterization for in 1478 vivo locomotion of robotic capsule endoscopes: Experimental and modeling results. Journal 1479 of the Mechanical Behavior of Biomedical Materials. 2014 Nov;39:257–69.

1480 151. Dodou D, van den Berg M, van Gennip J, Breedveld P, Wieringa PA. Mucoadhesive films 1481 inside the colonic tube: performance in a three-dimensional world. Journal of The Royal 1482 Society Interface. 2008 Nov 6;5(28):1353–62.

1483 152. Dodou D, Girard D, Breedveld P, Wieringa PA. Intestinal locomotion by means of 1484 mucoadhesive films. In: ICAR ’05 Proceedings, 12th International Conference on 1485 Advanced Robotics, 2005 [Internet]. Seatle, WA, USA: IEEE; 2005 [cited 2019 Oct 4]. p. 1486 352–9. Available from: http://ieeexplore.ieee.org/document/1507435/

1487 153. Pensabene V, Valdastri P, Tognarelli S, Menciassi A, Arezzo A, Dario P. Mucoadhesive 1488 film for anchoring assistive surgical instruments in endoscopic surgery: in vivo assessment 1489 of deployment and attachment. Surgical Endoscopy. 2011 Sep;25(9):3071–9.

1490 154. Dodou D, Breedveld P, Wieringa PA. Stick, unstick, restick sticky films in the colon. 1491 Minimally Invasive Therapy & Allied Technologies. 2006 Jan;15(5):286–95.

1492 155. Sliker L, Ciuti G, E. Rentschler M, Menciassi A. Frictional resistance model for tissue- 1493 capsule endoscope Sliding contact in the gastrointestinal tract. Tribology International. 2016 1494 Jun 1;102.

1495 156. Kim JS, Sung IH, Kim YT, Kim DE, Jang YH. Analytical model development for the 1496 prediction of the frictional resistance of a capsule endoscope inside an intestine. Proceedings 1497 of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine. 1498 2007;221(8):837–845.

1499 157. Woo SH, Kim TW, Mohy-Ud-Din Z, Park IY, Cho J-H. Small intestinal model for 1500 electrically propelled capsule endoscopy. Biomedical engineering online. 2011;10(1):108.

1501 158. Baek NK, Sung IH, Kim DE. Frictional resistance characteristics of a capsule inside the 1502 intestine for microendoscope design. Proceedings of the Institution of Mechanical 1503 Engineers, Part H: Journal of Engineering in Medicine. 2004;218(3):193–201.

1504 159. Zhang C, Liu H, Tan R, Li H. Modeling of velocity-dependent frictional resistance of a 1505 capsule robot inside an intestine. Tribology Letters. 2012;47(2):295–301.

1506 160. Lyle AB, Luftig JT, Rentschler ME. A tribological investigation of the small bowel lumen 1507 surface. Tribology International. 2013 Jun;62:171–6.

1508 161. Lyle AB, Terry BS, Schoen JA, Rentschler ME. Preliminary Friction Force Measurements 1509 on Small Bowel Lumen When Eliminating Sled Edge Effects. Tribology Letters. 2013 1510 Sep;51(3):377–83.

1511 162. Kim J-S, Sung I-H, Kim Y-T, Kwon E-Y, Kim D-E, Jang YH. Experimental investigation 1512 of frictional and viscoelastic properties of intestine for microendoscope application. 1513 Tribology Letters. 2006 May;22(2):143–9.

1514 163. Wang KD, Yan GZ. Research on measurement and modeling of the gastro intestine’s 1515 frictional characteristics. Measurement Science and Technology. 2008;20(1):015803.

1516 164. Jiwoon Kwon, Sukho Park, Byungkyu Kim, Jong-Oh Park. Bio-Material Property 1517 Measurement System for Locomotive Mechanism in Gastro-Intestinal Tract. In: 1518 Proceedings of the 2005 IEEE International Conference on Robotics and Automation 1519 [Internet]. Barcelona, Spain: IEEE; 2005 [cited 2018 Nov 6]. p. 1303–8. Available from: 1520 http://ieeexplore.ieee.org/document/1570295/

1521 165. Wang X, Meng MQ. An experimental study of resistant properties of the small intestine for 1522 an active capsule endoscope. Proceedings of the Institution of Mechanical Engineers, Part 1523 H: Journal of Engineering in Medicine. 2010;224(1):107–118.

1524 166. Kwon J, Cheung E, Park S, Sitti M. Friction enhancement via micro-patterned wet elastomer 1525 adhesives on small intestinal surfaces. Biomed Mater. 2006;1(4):216.

1526 167. Xiaona Wang, Meng MQ-H. Study of Frictional Properties of the Small Intestine for Design 1527 of Active Capsule Endoscope. In: The First IEEE/RAS-EMBS International Conference on 1528 Biomedical Robotics and Biomechatronics, 2006 BioRob 2006 [Internet]. Pisa, Italy: IEEE; 1529 2006 [cited 2018 Nov 1]. p. 124–9. Available from: 1530 http://ieeexplore.ieee.org/document/1639071/

1531 168. Bai JPF, Burckart GJ, Mulberg AE. Literature Review of Gastrointestinal Physiology in the 1532 Elderly, in Pediatric Patients, and in Patients with Gastrointestinal Diseases. Journal of 1533 Pharmaceutical Sciences. 2016 Feb;105(2):476–83.

1534 169. Salminen S, Bouley C, Boutron M-C, Cummings JH, Franck A, Gibson GR, et al. 1535 Functional food science and gastrointestinal physiology and function. Br J Nutr. 1998 1536 Aug;80(S1):S147–71.

1537 170. Gajendran M, Loganathan P, Jimenez G, Catinella AP, Ng N, Umapathy C, et al. A 1538 comprehensive review and update on ulcerative colitis. Disease-a-Month. 2019 1539 Dec;65(12):100851.

1540 171. Effinger A, O’Driscoll CM, McAllister M, Fotaki N. Impact of gastrointestinal disease 1541 states on oral drug absorption - implications for formulation design - a PEARRL review. J 1542 Pharm Pharmacol. 2019 Apr;71(4):674–98.

1543 172. Sasaki Y, Hada R, Nakajima H, Fukuda S, Munakata A. Improved localizing method of 1544 radiopill in measurement of entire gastrointestinal pH profiles: colonic luminal pH in normal 1545 subjects and patients with Crohn’s disease. Am J Gastroenterol. 1997 Jan;92(1):114–8.

1546 173. Tojo R. Intestinal microbiota in health and disease: Role of bifidobacteria in gut 1547 homeostasis. WJG. 2014;20(41):15163.

1548 174. Liao Z, Liu H, Komaki R. Target Delineation for Esophageal Cancer. Journal of Women’s 1549 Imaging. 2003 Nov;5(4):177–86.

1550

1551

1552

1553

1554

1555

1556

1557

1558

1559

1560

1561

1562

1563

1564

1565

1566

1567

1568

1569 Declarations

1570 Ethics approval and consent to participate: Not applicable

1571 Consent for publication: Not applicable

1572 Availability of data and materials: Not applicable

1573 Competing interests: Not applicable

1574 Funding: Not applicable

1575 Authors' contributions:

1576 Lavinia Barducci, substantial contributions to the design of the work, to the writing and to

1577 the design of the figures

1578 Joseph C. Norton, contributions to the design and to the writing of the work

1579 Sunandita Sarker, substantial contribution to the writing of the work

1580 Sayeed Mohammed, contribution to the writing and to the design of the figures

1581 Ryan Jones, substantial contributions to the conception of the work and to the revision of

1582 the manuscript

1583 Pietro Valdastri, substantively revised the manuscript

1584 Benjamin Terry, substantial contributions to the conception of the work and to the revision

1585 of the manuscript.

1586 Acknowledgements: Not applicable

1587